U.S. patent application number 17/152429 was filed with the patent office on 2021-08-12 for systems and methods for automatically identifying and linking names in digital resources.
The applicant listed for this patent is NAMESFORLIFE, LLC. Invention is credited to George M. Garrity, Catherine M. Lyons, Charles T. Parker, Gerald P. Roston.
Application Number | 20210248204 17/152429 |
Document ID | / |
Family ID | 1000005542605 |
Filed Date | 2021-08-12 |
United States Patent
Application |
20210248204 |
Kind Code |
A1 |
Parker; Charles T. ; et
al. |
August 12, 2021 |
SYSTEMS AND METHODS FOR AUTOMATICALLY IDENTIFYING AND LINKING NAMES
IN DIGITAL RESOURCES
Abstract
The present invention provides systems and methods for
automatically identifying name-like-strings in digital resources,
matching these name-like-string against a set of names held in an
expertly curated database, and for those name-like-strings found in
said database, enhancing the content by associating additional
matter with the name, wherein said matter includes information
about the names that is held within said database and pointers to
other digital resources which include the same name and it
synonyms.
Inventors: |
Parker; Charles T.; (East
Lansing, MI) ; Lyons; Catherine M.; (Edinburgh,
GB) ; Roston; Gerald P.; (Saline, MI) ;
Garrity; George M.; (Okemos, MI) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
NAMESFORLIFE, LLC |
East Lansing |
MI |
US |
|
|
Family ID: |
1000005542605 |
Appl. No.: |
17/152429 |
Filed: |
January 19, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
16272783 |
Feb 11, 2019 |
10896236 |
|
|
17152429 |
|
|
|
|
15615171 |
Jun 6, 2017 |
10204168 |
|
|
16272783 |
|
|
|
|
12685964 |
Jan 12, 2010 |
9672293 |
|
|
15615171 |
|
|
|
|
61143986 |
Jan 12, 2009 |
|
|
|
61184187 |
Jun 4, 2009 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06F 16/955 20190101;
G06F 16/9566 20190101; G06F 16/3331 20190101; G06F 16/90344
20190101; G06F 16/951 20190101 |
International
Class: |
G06F 16/955 20060101
G06F016/955; G06F 16/951 20060101 G06F016/951 |
Goverment Interests
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
[0002] This invention was made with government support under STIR
contract number DE-FG02-07ER86321 awarded by Department of Energy.
The United States government has certain rights in the invention.
Claims
1. A method for adding tags to a first digital resource, the method
comprising: creating a second digital resource which comprises the
content of the first digital resource; finding within said second
digital resource at least one occurrence of a name-like-string,
wherein said name-like-string is contained within a database, said
database relates said name-like-string to other name-like-strings
by means of a taxonomic structure based on rules of nomenclature,
and said database is expertly curated; obtaining from said database
for said name-like-string a globally unique identifier associated
with said string; and inserting a tag with said identifier into
said second digital resource.
2.-37. (canceled)
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a Continuation of and claims the benefit
of U.S. Non-Provisional application Ser. No. 15/615,171, filed Jun.
6, 2017, which is a Continuation of U.S. Non-Provisional
application Ser. No. 12/685,964, filed Jan. 12, 2010, now U.S. Pat.
No. 9,672,293, issued Jun. 6, 2017, which claims the benefit of
U.S. Provisional Application No. 61/143,986, filed Jan. 12, 2009,
and U.S. Provisional Application No. 61/184,187, filed Jun. 4,
2009, each of which are incorporated herein by reference in their
entireties.
FIELD OF THE INVENTION
[0003] The present invention relates to systems and methods for
automatically identifying and tagging name-like-strings in digital
resources.
BACKGROUND OF THE INVENTION
[0004] Using biological entities as an example, it is known that
the simple act of naming an entity, which is part of a large,
complex classification or taxonomic system, has potentially
far-reaching and long-lived consequences. For example, those names
ascribed to organisms serve as a primary entry point into the
scientific, medical, and technical literature and figure
prominently in countless laws and regulations governing various
aspects of commerce, public safety and public health. Biological
names also serve as a primary entry point into many of the central
databases that the scientific community and the general public rely
on. While legalistic Codes of Nomenclature exist that govern the
formation, assignment, and usage of names to biological taxa, the
process of biological classification is not governed by these Codes
of Nomenclature. Taxonomies represent the scientific opinions of
the individuals who create them, and may be of varying quality or
consistency. Hence, legitimate and valid names may be ascribed to
poorly-formed taxa and illegitimate and invalid names may be
assigned to well-formed and/or correctly identified taxa. Moreover,
biological names are neither unique nor permanent. A single
organism can bear multiple names (synonyms) that represent
differing taxonomic opinions that may have been rendered either in
sequence or in parallel. Instances of homonymy also occur, in which
a single name may refer to more than one group of organisms that
are of markedly different evolutionary lineages (e.g. the genus
name Bacillus applies to bacteria and insects). Orthographic
variants may also occur, arising from correction of nomenclatural
errors.
[0005] Those whose work involves bacteria, from research scientists
and clinicians to public officials and bioterrorism experts, face
an ever-growing list of names for everything from whole organisms
to individual genes, proteins, and sub-cellular components. While
many bacteria are harmless and others can be beneficial, some are
the causative agents of potentially lethal diseases. Knowing which
is which requires reasoned judgment and on-demand access to the
correct information, information that is typically accessed using a
scientific name. With rapid advances in knowledge of the depth and
breadth of bacterial diversity, the list of names not only grows,
it also undergoes incremental re-definition on a daily basis.
[0006] While these additions and name changes are of considerable
interest to a relatively small number of experts engrossed in
bacterial classification, they present a significant problem to
both end-users and information providers, who must invest a
significant and increasing effort to make the connections between
new names to names in publications that predate any change. Failure
to maintain name currency may prevent scientists and non-scientists
alike from tracking important developments in their field and may
trigger inappropriate or life-threatening responses.
[0007] Simply put, the continual rapid change in the nomenclature
of bacteria has outstripped the ability of most end-users,
information providers, and device manufacturers to keep pace. What
is needed is a bacterial name-lookup service that permits on-demand
access to the correct information at the point of need, regardless
of the name that is used. Ideally, that same service would provide
a means of synchronizing updates across the entire field, in the
scientific literature and databases, in diagnostic instruments, and
wherever else bacterial names must be resolved.
[0008] The disjunction between nomenclature and taxonomy leads to
an accumulation of dubious names in the literature and databases.
While experts in taxonomy and biological nomenclature may be able
to recognize and correctly interpret such circumstances, few others
have the requisite skills to do so, resulting in frequent
misapplication of names and misinterpretation of the taxonomic
record. From a practical, legal, or regulatory sense, either
incorrect nomenclature or errors in classification or
identification can have significant and unintended consequences.
For example, these errors may lead to the addition or removal of
biological species from lists of tightly-regulated organisms such
as those appearing on the CDC list of Restricted Select Agents,
those governed by the USDA APHIS program, those covered by the
Endangered Species Act, or those restricted by packaging and
shipping regulations maintained by the USPS, DOT and various
international bodies such as the International Airline Transport
Association (IATA). The use of biological names as a means of
information retrieval is not reliable as these names are neither
unique nor persistent.
[0009] The present invention provides systems and methods for
automatically enabling digital resources, based on an underlying
information architecture, as well as a means of accessing data
related to those entities in a networked environment using
persistent, globally unique identifiers.
[0010] The invention enables efficient mining of enormous amounts
of biological literature and annotations relevant to a particular
organism, even if the organism has undergone multiple taxonomic
re-classifications and renaming. This will prevent the introduction
of errors related to semantic inconsistencies as well as loss of
information. The present invention "future-proofs" those products
to which it is applied by providing a method of persistently
linking biological names or other terms found in published content
to the expertly managed information about the origins, definition,
and current usage of those terms. The present invention can also
serve as a fine-grained marketing tool by establishing persistent
links between terms and related products for sale. The present
invention provides services which assure content providers that
their offerings will always be accessible to their end-users and
eliminate the burden and costs of tracking and managing rapidly
changing biological terminologies.
SUMMARY OF THE INVENTION
[0011] In one embodiment the invention is a software method which
includes a stand-alone program that operates on prepublication
materials, and in another embodiment a browser-based extension that
operates on the content of a web-page. The invention also includes
methods for generating revenue based on the additional matter used
to enhance digital resources.
[0012] In another embodiment, the present invention provides
systems and methods for automatically identifying name-like-strings
in digital resources, matching these name-like-string against a set
of names held in an expertly curated database, and for those
name-like-strings found in said database, enhancing the content by
associating additional matter with the name, wherein said matter
includes information about the names that is held within said
database and pointers to other digital resources which include the
same name and its synonyms.
[0013] The invention allows rapid incorporation of changes to keep
pace with advances in the field. In one embodiment, the invention
uses technology which includes an underlying structure for
classifying types of biological data; persistent identifiers for
each term; and a database model based on the rules of biological
nomenclature. In various embodiments the invention uses Digital
Object Identifiers (DOI.RTM.), which guarantees that links are
persistent, to act as surrogates for names and provide a direct
path to the most current name. Along with each name, the invention
in certain embodiments stores both attributes and methods
associated with the name. The attributes may include specific
information that is important for identifying the object, such as
the date of discovery, the name of the discoverer, etc. The manner
in which these objects are related to each other forms the basis
for ensuring that the meaning of the objects remains consistent
over time. A list of services, many of which can be monetized, may
also be associated with the specific object. In the absence of the
inventive technology and services, content providers would either
have to bear the costs of updating their publications or
information products manually at regular intervals or else risk a
rapid decay in their investment.
[0014] In some embodiments, the invention provides a method for
adding a tag to a digital resource. The method comprises finding
within said digital resource at least one occurrence of a
name-like-string, wherein said name-like-string is contained within
a database, and said database relates said name-like-string to
other name-like-strings by means of a taxonomic structure based on
rules of nomenclature, and said database is expertly curated. A
globally unique identifier is obtained from said database based
upon said name-like-string. A tag comprising said identifier is
inserted into said digital resource. The method may be implemented
by a computer.
[0015] In this way, a plurality of related name-like-strings may be
associated with a single globally unique identifier. As such, a
plurality of digital resources including different, but related,
name-like-strings maybe processed so that a common globally unique
identifier is inserted into each of the plurality of digital
resources. Where, for example, a single organism is identified
using different name-like-strings in different resources, it is
therefore possible to identify that the different name-like-strings
identify the same organism by insertion of the globally unique
identifier in each of the digital resources.
[0016] A method for adding a tag to a digital resource. The method
includes finding within the digital resource at least one
occurrence of a name-like-string, wherein the name-like-string is
contained within a database. The database relates the
name-like-string to other name-like-strings by means of a taxonomic
structure based on rules of nomenclature, where the database is
expertly curated. The method also includes obtaining from the
database a globally unique identifier with the name-like-string and
modifying the digital resource by inserting a tag with the
identifier.
[0017] A method for adding content to a database of
name-like-strings. The method includes finding within a digital
resource at least one occurrence of a name-like-string, wherein the
name-like-string is not contained within the database. The database
relates the name-like-strings to other name-like-strings by means
of a taxonomic structure based on rules of nomenclature, where the
database is expertly curated. The method also includes having an
expert ensure that the name-like-string is in fact a valid name and
creating a new record within the database comprising the
name-like-string, a content identifier, and named relations to
other name-like-strings already within the database.
[0018] A method for providing information and services to End-Users
from digital content having at least one content identifier
embedded therein. The method includes accessing an expertly curated
database which relates name-like-strings to other name-like-strings
by means of a taxonomic structure based on rules of nomenclature
and using the content identifier to obtain information about a
related name-like-string existing in the digital content. The
method also includes accessing a database which associates
name-like-strings to services and information and using the content
identifier to obtain the services and information about the related
name-like-string existing in the digital content. The method
further includes displaying information and providing a means for a
user to select from the information displayed.
[0019] A computer program product including a computer usable
medium having a computer readable program code embodied therein,
where the computer readable program code is adapted to be executed
to implement a method for adding tags to a first digital resource.
The method includes creating a second digital resource which
includes the content of the first digital resource and finding
within the second digital resource at least one occurrence of a
name-like-string, wherein the name-like-string is contained within
a database, the database relates the name-like-string to other
name-like-strings by means of a taxonomic structure based on rules
of nomenclature, and the database is expertly curated. The method
also includes obtaining from the database for the name-like-string
a globally unique identifier associated with the string and
inserting a tag with the identifier into the second digital
resource.
[0020] A computer program product including a computer usable
medium having a computer readable program code embodied therein,
where the computer readable program code is adapted to be executed
to implement a method for adding a tag to a digital resource. The
method includes finding within the digital resource at least one
occurrence of a name-like-string, wherein the name-like-string is
contained within a database, the database relates the
name-like-string to other name-like-strings by means of a taxonomic
structure based on rules of nomenclature, and the database is
expertly curated. The method also includes obtaining from the
database a globally unique identifier uniquely associated with the
name-like-string and modifying the digital resource by inserting a
tag with the identifier.
[0021] A computer program product including a computer usable
medium having a computer readable program code embodied therein,
the computer readable program code adapted to be executed to
implement a method for adding content to a database of
name-like-strings. The method includes finding within a digital
resource at least one occurrence of a name-like-string, wherein the
name-like-string is not contained within the database, the database
relates the name-like-strings to other name-like-strings by means
of a taxonomic structure based on rules of nomenclature, and the
database is expertly curated. The method also includes having an
expert ensure that the name-like-string is in fact a valid name and
creating a new record within the database comprising the
name-like-string, a content identifier, and named relations to
other name-like-strings already within the database.
[0022] A computer program product including a computer usable
medium having a computer readable program code embodied therein,
the computer readable program code adapted to be executed to
implement a method for providing information and services to
End-Users from digital content having at least one content
identifier embedded therein. The method includes accessing an
expertly curated database which relates name-like-strings to other
name-like-strings by means of a taxonomic structure based on rules
of nomenclature and using the content identifier to obtain
information about a related name-like-string existing in the
digital content. The method also includes accessing a database
which associates name-like-strings to services and information and
using the content identifier to obtain the services and information
about the related name-like-string existing in the digital content.
The method further includes displaying information and providing a
means for a user to select from the information displayed.
[0023] A method for building a graph relating digital resources.
The method includes calculating a distance between two digital
resources using a mathematical formula that is based upon the
number of occurrences of a name in each of the digital
resources.
[0024] A method for tracking a number of occurrences of a
name-like-string in a digital resource. The method includes
counting a first number of occurrences of the name-like-string in a
digital resource for a first time period, and calculating a first
average and a first standard deviation of the first number of
occurrences for the first time period.
[0025] A method for generating a visualization of taxonomic data.
The method includes obtaining a selected name object and an
ordering basis from a user; searching a database for name objects
related to the selected name object; retrieving one or more related
name objects from the database; extracting time-based publication
information for the selected name object and each of the one or
more related name objects retrieved from the database; and
determining a relative arrangement of the selected name object and
the one or more related name objects retrieved from the database
using the ordering basis obtained from the user, thereby producing
an ordered list of name objects. The method also includes
generating a first output image comprising a graph, the graph
including the selected name object and the one or more related name
objects retrieved from the database, the graph further including a
time-based axis, such that each of the selected and related name
objects is represented on the graph by a first visual indicator
which depicts the time-based publication information extracted for
each respective name object; and presenting the first output image
to the user.
DESCRIPTION OF THE DRAWINGS
[0026] The above, as well as other advantages of the present
disclosure, will become readily apparent to those skilled in the
art from the following detailed description, particularly when
considered in the light of the drawings described herein.
[0027] FIG. 1 shows the problem faced by End-Users without the
benefit of the present invention.
[0028] FIG. 2 shows how the present invention mitigates the current
problem by facilitating access to other documents that contain
verified references to a particular object.
[0029] FIG. 3 shows the business process by which the actionable
services associated with enhanced objects are employed.
[0030] FIG. 4 shows a network diagram with all relevant computers
identified.
[0031] FIG. 5 shows how additional metadata may be requested from a
server and be displayed within the context of an enhanced
document.
[0032] FIG. 6 is a view of a document as it exists prior to
enhancement.
[0033] FIG. 7 is a view of the document enhanced with the presently
described invention, where the name-like-strings have been made
visually distinct.
[0034] FIG. 8 is a view of the document after an End-User has
activated the event handling code which is included within the
enhancement of the digital resource.
[0035] FIG. 9 shows the Workflow model for situations in which the
original content is read-only and which results in the production
of a new enabled document. This process would typically be
integrated into an existing publication process.
[0036] FIG. 10 shows the Plug-in model for those situations where
the digital resource resides in memory and is enhanced without
generating an entirely new document. This process would typically
be implemented as an add-on, extension, or plug-in for a web
browser, document editor, or similar extensible application.
[0037] FIG. 11 shows the general process for enabling the content
of a digital resource.
[0038] FIG. 12 shows examples of name-like-strings as they might
appear in a digital resource, and how these might be recognized as
name-like-strings based on at least one of typographic, formatting,
and stylistic conventions.
[0039] FIG. 13 shows an embodiment of a process for extracting
name-like strings from a digital resource.
[0040] FIG. 14 shows the process performed to verify that
name-like-strings are in fact names.
[0041] FIG. 15 shows the a high-level overview of the database
curation process.
[0042] FIG. 16 shows an example of a time-based visualization
method.
[0043] FIG. 17 shows another example of a time-based visualization
method.
[0044] FIG. 18 shows yet another example of a time-based
visualization method.
[0045] FIG. 19 shows an example of a taxonomic-based visualization
method at a the point in time indicated in FIG. 18.
[0046] FIG. 20 shows a flow chart of a method for creating a
time-based visualization.
[0047] FIG. 21 shows a flow chart of a method for creating a
taxonomic-based visualization.
[0048] FIG. 22 shows an example of building a graph to identify the
relatedness of a set of digital resources.
[0049] FIG. 22A shows an example of six digital resources, each
with names.
[0050] FIG. 22B shows the calculated distance between the example
digital resources.
[0051] FIG. 22C shows a network diagram of the example digital
resources.
[0052] FIG. 23 shows a diagram of a computer that can be used to
implement embodiments of the invention.
DEFINITIONS
[0053] To facilitate an understanding of the present invention, a
number of terms and phrases are defined below:
[0054] N4L refers to an application of an information architecture
in some embodiments of the present invention that disambiguates
scientific names and taxonomic opinions, wherein ambiguity in the
meaning and context of scientific names and taxonomic opinions is
resolved through the separation and association of the following
entity types: Name, Taxon, Nomos, Practitioner, and Exemplar.
Relations between Names are resolved through analysis of naming
Events. A naming Event is a kind of nomenclatural event (there are
other kinds) that names a Taxon in a given state, but not in
perpetuity, and optionally modifies the status of a name created in
a previous event. Where a nomenclatural event refers directly or
indirectly to a previous event, the event points to the previous
event and to corresponding entities. Referring events include but
are not limited to: emended naming, the naming of a later synonym
that changes the synonym type of the name referred to, and all
non-naming events. Naming events may be used as the basis for the
schema of a database or an XML/SGML repository.
[0055] In the application, the entities are represented by means of
persistent, uniquely identified, addressable, structured documents,
referred to as "information objects."
[0056] In the application, information objects are used in the
construction of nomenclatural and taxonomic "services."
[0057] Information Objects:
[0058] An information object (also referred to as a N4L object) is
a virtual representation of an "entity"; it may comprise
descriptive text, metadata, and data. Each information object is
identified by a persistent, globally unique identifier and is
directly addressable. Multiple persistent, globally unique
identifiers can be employed to identify an information object so
long as each derives from separate family of identifiers, e.g., an
information object may be identified by one DOI, one LSID, etc. One
type of information object is a name object.
[0059] In various embodiments, N4L information objects are
structured and interlinked. It is therefore possible to build
services based on preset queries. Such services can retrieve and
display information from N4L objects, and also use information
drawn from N4L objects to link to other resources on the
Internet.
[0060] Entity:
[0061] N4L entities are entities that exist in the physical or
constructed world.
[0062] A "Practitioner" is a scientist or other person who asserts
a taxonomic opinion. In some embodiments, the information object or
other information associated with a Practitioner contains a
reliability score computed from other information objects that
accounts for historic performance of the Practitioner.
[0063] An "Exemplar" is a curated whole or partial organism, or a
representation thereof.
[0064] A "Taxon" is a group of one or more Exemplars or a group of
one or more subordinate Taxa (plural form of Taxon) that are
circumscribed by a Practitioner according to a Nomos. Exemplars and
subordinate Taxa can belong to more than one Taxon simultaneously,
as long as the Taxa they belong to are revealed by different Nomoi
(plural form of Nomos).
[0065] A "Nomos" is a methodology (e.g. heuristics, algorithms or
principles) used by a Practitioner to interpret data and to reveal
and circumscribe Taxa. In one embodiment, an analysis of 16S rRNA
gene sequence similarity forms the basis of a Nomos. In some
embodiments, phylogenetic, phenotypic, genotypic, phenetic, genomic
or polyphasic approaches form the basis of multiple Nomoi. As
sequence data accumulates for multiple loci, taxonomic
methodologies will become differentiated, and Nomoi will provide a
means of formal distinction.
[0066] A "Name" is a record of an event, by means of a dated
publication, in which a Practitioner circumscribes a Taxon. In one
embodiment, a Name is `new`, `emended`, `corrected`, `rejected`,
`conserved`, or `combined`. Names have properties that can be
modified by later events.
[0067] In one embodiment, "Name" refers to an archived
bibliographic record, which links to two other objects: a Taxon
object and a publication object. For an Emended Name or a New
Combination, links are made both to the current state of the Taxon
and to the Taxon as it was when the name was a `new` Name. Once a
Name is recorded, it will persist unchanged, aside from its synonym
type.
[0068] A "name-like-string" refers to a noun or compound noun
phrase that may appear to a parser or a human as a reference to a
biological or scientific name or concept. It may appear in special
formatting, in full or abbreviated form, and upon first usage
points to a published definition of the concept and/or object to
which the name applies.
[0069] An "Annotation" is a published comment, under strict
editorial control, which can be added by a Practitioner, an
End-User and/or a skilled N4L curator.
[0070] A "taxonomy" is a particular classification of objects
arranged in a hierarchical structure. Taxonomies are generally
organized by supertype-subtype relationships, wherein by definition
the subtype (child) has the same properties as the supertype
(parent) plus one or more additional properties. Objects at the
same level of the taxonomic hierarchy can be said to have the same
"taxonomic rank".
[0071] One skilled in the art will recognize that biological
exemplars, taxa, and names represent but one instance of a taxonomy
of objects related by nomoi. The present invention applies equally
to all other sets of exemplars, taxa, and names related by nomoi
wherein said nomoi are prescribed and the relationship among the
entities is maintained expertly.
[0072] Services:
[0073] Since N4L objects are directly addressable, they can be
visited like any other web page. Information objects, and services
derived from them, can be accessed from within other "digital
resources" and provided in human readable form, machine readable
form, or both.
[0074] In some embodiments, services are implemented using standard
technology, including DOI (Digital Object Identifier, a proprietary
form of Handle resolution services deployed by the Publishers
Interlinking Association (PILA) under the governance of the
International DOI Foundation (IDF)) multiple resolution in one
embodiment, and web services in other embodiments. In yet another
embodiment, resolution of a persistent identifier to services
results in the rendering of graphical objects in the user
interface, enabling the user to select and execute the chosen
service. Graphical objects include text-based dropdown menus and
pop-up windows, and interactive graphs (e.g., 2D, 3D, and
multi-dimensional plots, tree graphs) containing active regions
that respond to user behavior (e.g., the rollover of a mouse) by
offering or executing services.
[0075] N4L services enable prospective survey of Names, Taxa, and
Exemplars: a reference to an N4L object enables the retrieval not
only of information that was current when the reference was made,
but also of information updates made since then. N4L services
consider a Name to be a formal reference to an archived event in
the time continuum, and therefore to be an index to all related
events.
[0076] N4L services enable detailed comparative analysis. The
reification of Nomos as an object distinct from the Taxa it
reveals, forces the separation of partially or completely
overlapping taxa that arise from the application of different
taxonomic methods, thereby clarifying differing taxonomic opinions.
Revision of an existing Taxon as a result of new data is distinct
from assertion of a Taxon that is different from an existing one,
though overlapping. The former case is a comparison of two states
of the same Taxon. The latter case is a comparison of two
heteronomic Taxa (which we define as overlapping Taxa from
different Nomoi). Tools for taxonomic comparison (e.g., those
provided by the present invention) become increasingly important as
taxonomic methodology increases in complexity and becomes more
differentiated.
[0077] N4L services are provided by computer programs, or
extensions, which access the N4L database and/or the N4L-CI
(contextual index) database. The N4L database maintains the
relationship between the information objects, i.e., it is the
instantiation of the taxonomic record. The N4L-CI database provides
a many-to-many relationship among the information objects and the
digital resources in which they appear, along with other
metadata.
[0078] Digital Resource
[0079] A digital resource is any computer-storable item of textual,
data, graphical, or multimedia content, such as STM (scientific,
technical and medical) literature, patents, BRC (Biological
Resource Center) databases, legal literature, diagnostic machinery
databases, contracts, treaties, material transfer agreements, prior
informed consent agreements, certificates of origin, other
documents establishing provenance, bills of lading, shipping
documents, specification sheets, regulations or laws as applied to
holding, transportation, or working with an organism, and
microarrays/underlying databases which are provided by publishers,
culture collections, herbaria, museums, biological databases,
governments, NGOs (Non-Governmental Organizations) and others. A
digital resource may include encoded text, e.g. in ASCII format, in
which case the text may be analyzed directly for occurrences of
name-like-strings. In other instances, a digital resource may
include images containing text, in which case the text data may
have to be extracted from the images as an initial step, for
example using optical character recognition (OCR), before the text
can be analyzed for occurrences of name-like-strings.
[0080] Collection Description
[0081] Biological taxonomies are grounded in the typological
concept. For each formally named taxon, there exists one or more
Exemplars. In instances in which multiple Exemplars exist, one
Exemplar is designated the type, the holotype, or the lectotype,
and it serves as the principal reference object against which
Practitioners make their comparisons. Exemplars are held in various
collections and said collections may be housed in museums,
herbaria, botanical or zoological gardens, academic institutions,
culture collections or biological resource centers. The form of
Exemplars is governed by the different Codes of Nomenclature,
varies by discipline and may be either preserved or living, whole
organisms, parts of whole organisms or pictorial or written
representations. Sequences of one or more genes or entire genomes
of Exemplars may serve as supporting data and may, in the future,
constitute an alternative form. Each Exemplar (type specimen) is
uniquely identified with one or more collection identifiers (e.g.,
accession number) as well as by those identifiers ascribed to the
material by the Practitioner, prior to deposit in the said
collection(s). The citation of these identifiers in formal
taxonomic descriptions is governed by the applicable Code and
provides a means by which Practitioners may locate and retrieve
Exemplars.
[0082] In the N4L model, Exemplars are aggregating objects that
contain information about organisms, especially those from which
biological material is curated; for example the type or a non-type
strain/specimen and associated data and services specific to that
strain/specimen. Exemplars are assigned a unique and persistent
identifier. Where the same organism is deposited in multiple
collections, by means of cloning, serial passage or segmentation,
references to those multiple deposits are held in a single Exemplar
object. Exemplars are information objects that record their
relationships with each other, with N4L information objects of
other classes (e.g., Name, Taxon) and with other information
resources. Referencing deposits and data by means of N4L Exemplars
alleviates much of the current ambiguity associated with
determining equivalency and availability of such materials
[0083] Globally Unique Identifiers and Resolution Systems
[0084] A `globally unique identifier` is a type of identifier used
in software applications to provide a reference which is guaranteed
to be unique in any context. One method for producing these
identifiers has been standardized by the Open Software Foundation
(OSF) and documented as part of ISO/IEC 11578:1996 "Information
technology--Open Systems Interconnection--Remote Procedure Call
(RPC)" and more recently in ITU-T Rec. X.667 | ISO/IEC 9834-8:2005.
This standard refers to the identifiers as Universally Unique
Identifiers (UUIDs). A common implementation of this standard has
been promulgated by Microsoft, which refers to these identifiers as
Globally Unique Identifiers (GUIDs).
[0085] While globally unique identifiers work well within the
context of a computer, or even a network of computers, it has been
recognized that there needed to be implemented methods which ensure
that the object pointed to by such an identifier always resolves to
the object being identified. This need led to the creation of
`persistent identifiers`, globally unique identifiers provided by
an organization which guarantees that the identifier will resolve
to the object identified in perpetuity. This guarantee has two
components--a `social` contract and a resolution system.
[0086] In the scientific, informatics, and publishing communities,
resolution software and services are used to provide persistent
access to (especially) distributed resources. Resolution systems
that find use in an embodiment of N4L are typically Uniform
Resource Identifiers (URIs), and include Handles and Uniform
Resource Names (URNs). Digital Object Identifiers (DOI), which are
an implementation of Handle, and Life Science Identifiers (LSID),
which are an implementation of URN, are both good examples.
Archival Resource Keys (ARK) and Persistent Uniform Resource
Locators (PURL) are other resolution systems. More than one type of
persistent identifier/resolution system could be used in an
implementation of a N4L application, depending on the needs of the
community that is making use of the entities represented by the
information objects.
General Definitions
[0087] As used herein, the terms "processor" and "central
processing unit" or "CPU" are used interchangeably and refer to a
device that is able to read a program from a computer memory (e.g.,
ROM or other computer memory) and perform a set of steps according
to the program.
[0088] As used herein, the terms "computer memory" and "computer
memory device" refer to any storage medium readable by a computer
processor. Examples of computer memory include, but are not limited
to, RAM, ROM, computer chips, digital video discs (DVD), compact
discs (CDs), hard disk drives (HDD), and magnetic tape.
[0089] As used herein, the term "computer readable medium" refers
to any device or system for storing and providing information
(e.g., data and instructions) to a computer processor. Examples of
computer readable media include, but are not limited to, DVDs, CDs,
hard disk drives, magnetic tape and servers for streaming media
over networks. In various embodiments, aspects of the present
invention including data structures and methods may be stored on a
computer readable medium.
[0090] As used herein, the term "encode" refers to the process of
converting one type of information or signal into a different type
of information or signal to, for example, facilitate the
transmission and/or interpretability of the information or signal.
For example, image files can be converted into (i.e., encoded by)
electrical or digital information. Likewise, light patterns can be
converted into electrical or digital information that provides an
encoded video capture of the light patterns.
[0091] As used herein, the term "N4L enhancement" refers to
metadata embedded within a digital resource which responds to a
user-initiated behavior in a predictable manner. For example, when
a hyperlink in a HTML document is selected by a user's mouse, the
web browser loads the page referenced. More sophisticated behaviors
can occur with the use of, for example, a JavaScript event
model.
[0092] As used herein, the term "internet" refers to any collection
of networks using standard protocols. For example, the term
includes a collection of interconnected (public and/or private)
networks that are linked together by a set of standard protocols
(such as TCP/IP, HTTP, and FTP) to form a global, distributed
network. While this term is intended to refer to what is now
commonly known as the Internet, it is also intended to encompass
variations that may be made in the future, including changes and
additions to existing standard protocols or integration with other
media (e.g., television, radio, etc). The term is also intended to
encompass non-public networks such as private (e.g., corporate)
intranets and the private military internets.
[0093] As used herein, the terms "World Wide Web" or "web" refer
generally to both (i) a distributed collection of interlinked,
user-viewable hypertext documents (commonly referred to as Web
documents or Web pages) that are accessible via the Internet, and
(ii) the client and server software components which provide user
access to such documents using standardized Internet protocols.
Currently, the primary standard protocol for allowing applications
to locate and acquire Web documents is HTTP, and the Web pages are
encoded using HTML. However, the terms "Web" and "World Wide Web"
are intended to encompass future markup languages and transport
protocols that may be used in place of (or in addition to) HTML and
HTTP.
[0094] As used herein, the term "web site" refers to a computer
system that delivers content and services over a network using the
standard protocols of the World Wide Web. Typically, a Web site
corresponds to a particular Internet domain name and includes the
content associated with a particular organization. As used herein,
the term is generally intended to encompass both (i) the
hardware/software server components that delivers the content and
services over the network, and (ii) the "back end"
hardware/software components, including any non-standard or
specialized components, that interact with the server components to
perform services for web site users and the tools used by said web
site users.
[0095] As used herein, the term "HTML" refers to Hyper-Text Markup
Language that is a standard coding convention and set of codes for
attaching presentation and linking attributes to informational
content within documents. During a document authoring stage, the
HTML codes (referred to as "tags") are embedded within the
informational content of the document. When the Web document (or
HTML document) is subsequently transferred from a Web server to a
browser, the codes are interpreted by the browser and used to parse
and display the document. Additionally, in specifying how the Web
browser is to display the document, HTML tags can be used to create
links to other Web documents (commonly referred to as
"hyperlinks").
[0096] As used herein, the term "HTTP" refers to Hyper-Text
Transport Protocol that is the standard World Wide Web
client-server protocol used for the exchange of information (such
as HTML documents, and client requests for such documents) between
a browser and a Web server. HTTP includes a number of different
types of messages that can be sent from the client to the server to
request different types of server actions. For example, a "GET"
message, which has the format GET, causes the server to return the
document or file located at the specified URL.
[0097] As used herein, the term "URL" refers to Uniform Resource
Locator that is a unique address that fully specifies the location
of a file or other resource on the Internet. The general format of
a URL is protocol://machine address:port/path/filename. The port
specification is optional, and if none is entered by the user, the
browser defaults to the standard port for whatever service is
specified as the protocol. For example, if HTTP is specified as the
protocol, the browser will use the HTTP default port of 80.
[0098] As used herein, the term "in electronic communication"
refers to electrical devices (e.g., computers, processors, etc.)
that are configured to communicate with one another through direct
or indirect signaling. For example, a conference bridge that is
connected to a processor through a cable or wire, such that
information can pass between the conference bridge and the
processor, are in electronic communication with one another.
Likewise, a computer configured to transmit (e.g., through cables,
wires, infrared signals, telephone lines, etc.) information to
another computer or device, is in electronic communication with the
other computer or device.
[0099] As used herein, the term "web service" is a software system
designed to support interoperable machine-to-machine interaction
over a network. A specific web service would consist of an
Application Programming Interface (API) that may be invoked from a
client system to retrieve data or execute a process on a remote
system.
[0100] As used herein, the term "transmitting" refers to the
movement of information (e.g., data) from one location to another
(e.g., from one device to another) using any suitable means.
[0101] A "tag" is a way of electronically marking content to
associate it with additional information and metadata, or to impose
a specific structure on the specific content to make it distinct
from the surrounding content. Examples of tags include XML and HTML
elements. One type of tag is a metadata tag.
[0102] As used herein, "metadata" refers to descriptive data about
data of any type. Metadata may describe a single datum or entire
collections of similar data. It provides context for data and
additional information to the user, and may associate one piece of
data with another piece of data based on properties shared in
common. Metadata often refers to the structure (also referred to as
the schema) of a database or of a document encoded in XML or SGML.
In one embodiment, metadata refers to the description of the types
of data contained in Name, Taxon, Exemplar, Nomos and Practitioner
information objects and other types of structured data that may be
linked to such information objects.
[0103] As used herein, the term "XML" refers to Extensible Markup
Language, an application profile that, like HTML, is based on SGML
(Standard Generalized Markup Language). XML differs from HTML in
that: information providers can define new tag and attribute names
at will; document structures can be nested to any level of
complexity; any XML document can contain an optional description of
its grammar for use by applications that need to perform structural
validation. XML documents are made up of storage units called
elements which contain either parsed or unparsed data. Parsed data
is made up of characters, some of which form character data, and
some of which form markup. Markup encodes a description of the
document's storage layout and logical structure. XML provides a
mechanism to impose constraints on the storage layout and logical
structure, to define constraints on the logical structure and to
support the use of predefined storage units. A software module
called an XML processor is used to read XML documents and provide
access to their content and structure.
[0104] As used herein, the term "interactive graphic", refers to a
digitized image representing the relationship(s) among two or more
independent and dependent variables in which the labels are
available for processing by an N4L application or can directly link
to N4L information objects directly. In one embodiment, such
interactive graphics include, but are not limited to heatmaps,
phylogenetic trees, scatter plots, line drawings or other figures
that use names or other labels that refer to organisms or features
of organisms that may be stored in Name, Taxon, or Exemplar
objects.
[0105] As used herein, the term "extension" refers to an auxiliary
computer program that interacts with a host application to provide
extended, specific capability. There exist a very large number of
extensible host applications, including web-browser applications,
such as Microsoft Internet Explorer and Mozilla Firefox; document
viewing programs, such as Adobe Acrobat; and document creation
programs, such as Microsoft Word. Extensions for these programs are
typically written in a high-level programming language using a
defined Application Programming Interface. Though subtly distinct
in actual application, as used herein, the term "plug-in" is used
synonymously with "extension" and "add-on".
[0106] As used herein, the term "string" refers to a sequence of
characters, wherein the characters may be encoded using any
recognized coding scheme, such as ASCII, Unicode, etc.
[0107] As used herein, the term "expertly curated database" refers
to a repository of domain-specific information that is maintained
under the guidance of one or more individuals having acknowledged
expertise in a particular field who are capable of ensuring the
accuracy and integrity of the information stored within said
repository. In one instance, an expertly curated database on the
nomenclature of Bacteria and Archaea is one that is maintained
under the guidance of one or more acknowledged experts who are
familiar with the appropriate codes of nomenclature that govern the
formation and application of biological names and are capable of
ensuring that the links among Names, Taxa, Exemplars, Nomi,
Practitioners and the taxonomic literature in which nomenclatural
events occur.
[0108] As used herein, "expert curation" is the process of
maintaining the integrity and accuracy of a database or other
knowledge resource by an acknowledged domain expert. In one
instance, expert curation refers to the maintenance of a database
or knowledge store of biological names by an acknowledged expert in
the field of biological nomenclature. In one embodiment, an expert
in the field of biological nomenclature is someone who holds a
post-graduate degree in biology. In another embodiment, an expert
in the field of biological nomenclature is someone who holds a
post-graduate degree in the biological sciences and specializes in
systemic biology. In yet another embodiment, an expert in the field
of biological nomenclature is someone who has authored or
co-authored at least one publication in the field of biological
nomenclature.
DETAILED DESCRIPTION OF THE INVENTION
[0109] The following description is merely exemplary in nature and
is not intended to limit the present disclosure, application, or
uses. It should also be understood that throughout the drawings,
corresponding reference numerals indicate like or corresponding
parts and features. In respect of the methods disclosed, the order
of the steps presented is exemplary in nature, and thus, is not
necessary or critical. In addition, while much of the present
invention is illustrated using biological taxonomy examples, the
present invention is not limited to these embodiments.
[0110] The present invention provides systems and methods for
automatically identifying name-like-strings in digital resources,
matching these name-like-strings against a set of names held in an
expertly curated database, and for those name-like-strings found in
said database, enabling said resource by associating a globally
unique identifier with said name, wherein said globally unique
identifier can be used to retrieve additional matter, such as
information about the name that is held within said database and
pointers to other digitally accessible resources which include the
same name or synonyms thereof, about said name.
[0111] The value of the invention is clearly demonstrated by FIG. 1
which highlights the difficulty experienced by an End-User without
the advantage of the presently described invention. An End-User
using a computer 16 is interested in learning about some named
organism 36, which is referenced in a journal article 34. The
journal article 34 is web-accessible since it resides on a web
server 32. Finding the journal article 34 potentially requires a
significant amount of effort on behalf of the End-User. Similarly,
a reference to the same organism 42, though quite possibly with a
different, but synonymous name, exists in another journal article
40 which is web-accessible via a web server 38. While mechanisms
using bibliographic referencing exist for linking the journal
article 40 to the journal article 34, these mechanisms do not
guarantee that the mention of the organism of interest appears in
the journal article 40. Furthermore, if the journal articles are
not linked via bibliographic reference, a similar degree of effort
that was needed to locate the journal article 34 is needed to
locate the journal article 40.
[0112] FIG. 2 demonstrates the value of the presently disclosed
invention. Display of the journal article 34 can be enhanced with a
browser plug-in as described below. During the course of processing
the journal article 34, the plug-in queries a N4L-CI server 14
which returns the location of the organism's name 42 in the journal
article 40. The N4L enhancement provides a direct link 44 from an
instance of the organism's name in the journal article 34 to an
instance of the same name 42 in the journal article 40. Thus, the
amount of time needed to locate successive references to the
organism of interest is reduced to essentially zero. It is noted
that in the context, the phrase "same name" also means any other
name or label that is known to denote an entity for which the N4L
database records a relation with the organism that is the focus of
research. The N4L database uses the expertly curated taxonomic
structure previously disclosed by Garrity et al. (US Pat. Appl.
Publ. No. 2005/0160059), incorporated by reference herein.
[0113] It should be apparent that the linking of journal articles
represents but one form of digital resources that may be enhanced
by the current invention. In addition to STM literature, patents,
BRC databases, legal literature, diagnostic machinery databases,
contracts, treaties, material transfer agreements, prior informed
consent, certificates of origin, other documents establishing
provenance, bills of lading, shipping documents, specification
sheets, and microarrays/underlying databases which are provided by
publishers, culture collections, herbaria, museums, biological
databases, governments, NGOs and others, can all benefit from
enhancement with the presently-described invention.
[0114] Embodiments of the invention are described in the following
non-limiting examples:
[0115] Example 1--shipping of biological materials: The regulations
relevant to the shipping of biological materials are based on the
name of the material being transported. The regulations are
formulated based on the nature of the material--those that are
deemed hazardous require additional protections, certifications,
etc., when transported from one location to another. Since this
system is based on names, at least two types of system failures are
possible. In a first failure case, the name of a hazardous material
could be changed such that the new name is no longer contained
within the list of hazardous materials. In this case, insufficient
precautions would be taken when shipping the material, thus
potentially putting peoples' lives at risk. In a second failure
case, the name of a benign material could be changed such that the
new name is within the list of hazardous materials. In this case,
extraordinary precautions would be taken when shipping the
material, thus placing unnecessary financial burden on the shipper.
In both cases, were the name in the regulation enhanced as
described herein, the regulation would be made independent of
changes to the material's name, thus the regulations would forever
dictate correct handling methods.
[0116] Example 2--patent searching: The number of patents which
reference biological names is quite large. For example, 375,000 US
patents issued between 1976 and 2007 were found to contain names of
bacteria. For inventors, patent attorneys, and patent examiners,
being able to quickly determine if a potential infringement exists
is vitally important. However, given the rate at which names
change, this becomes an onerous task. With the use of the presently
described invention, that which was onerous becomes trivial. All
other patents which include mention of an organism can be instantly
identified regardless of the name changes that may have occurred
subsequent to the publication of the patent.
[0117] FIG. 3 shows how a N4L application provides a means for
generating revenue. The N4L Database Maintainer has three primary
functions: to maintain the core taxonomic database (N4L), to
maintain the contextual index database (N4L-CI), and to provide
tools for enhancing digital resources.
[0118] Content Providers are typified by organizations which
produce articles, patents, regulations, etc. These organizations
can use the software tools developed in accordance with the present
invention to create enabled digital resources prior to their
publication (described in detail below). Using the present
invention, the process of enabling a digital resource makes it more
valuable because it ensures persistent access to proper names,
reduces the level of effort needed to validate a digital resource
prior to its publication, finds errors in a digital resource, and
facilitates the finding of additional digital resources. After
enablement, Content Providers make their content available to
End-Users, for example on a fee-per-use basis.
[0119] Service Providers are typified by organizations which
provide living matter, diagnoses based on sample inspection,
shipping services, etc. These organizations acquire tools from
NamesforLife and use these tools to enable digital resources prior
to their usage (described in detail below). Using the present
invention, the process of enabling a digital resource makes it more
valuable because it ensures persistent access to proper names,
ensures compliance with changing regulations, and provides an easy
means for End-Users to be aware of relevant products and services.
After enablement, the products and services offered by the Service
Providers are more easily found by End-Users, thereby potentially
increasing the sales of these products and services.
[0120] An End-User is an individual or organization who accesses
digital resources from Content Providers and/or uses products and
services offered by Service Providers. When accessing previously
enabled digital resources or when the digital resource is enabled
on-the-fly (described in detail below), software on the End-User's
computer enhances the digital resource by providing a menu of
information and/or actionable services associated with each
recognized name-like-string in the digital resource. Due to the
value of the information provided and/or due to the savings in
effort inherent with the use of the presently described invention,
the End-User will be willing to pay to use these services.
[0121] The following discussion shows how one embodiment of the
presently described invention may be implemented.
[0122] FIG. 4 depicts a network architecture with relevant
computers identified. The computer 16 provides access to networked
resources for an End-User. The End-User, working at a computer 16,
accesses certain information residing on a web server 10, typically
using a web browser. The present invention requires the use of a
software program, typically embodied as a N4L plug-in for said
web-browser.
[0123] A N4L Server 12 contains an expertly curated database which
relates information objects to each other based on taxonomic
structures which adhere to rules of nomenclature. The N4L-CI Server
14 includes a database that records a many-to-many relationship
between information objects and the digital resources in which they
are referenced. These records may also include additional metadata,
for example, the number of times the name (i.e., reference to an
information object) occurs within the digital resource. The N4L
Server 12 and The N4L-CI Server 14 may optionally be the same
computer and may optionally be connected to the same local area
network.
[0124] The functionality embodied in the N4L plug-in is shown in
the method 500 outlined in FIG. 5. The method 500 begins at start
505. Initially, the digital resource (e.g. web page document) is
examined to determine if it has been previously enabled 510. One
method for so doing is to check for the existence of a META tag
created during the publisher workflow process. A page is said to be
enabled if the name-like-strings within the page have associated
with them persistent identifiers which provide links to information
objects and additional services.
[0125] If the digital resource has not been previously enabled,
document content enablement 515 (described below in detail) occurs
on-the-fly. Once the resource is enabled, enhancement occurs. The
plug-in can embody the enhancement in at least one of three
manners: it may a priori retrieve metadata for all persistent
identifiers in the digital resource, it may retrieve metadata on
demand when the user selects a name-like-string for which there is
an associated persistent identifier, or it may use a combination of
the above methods.
[0126] The following steps are performed in a loop for each of the
name object metadata requested 520, after which the loop ends 540
and the method 500 ends 545. The persistent identifier is used to
make a query 525 against the N4L database 12. This query returns an
actionable code snippet, typically written in JavaScript, which
provides N4L enhancement of the name-like-string. Said N4L
enhancement may, for example, when chosen by the End-User, result
in the display of the chronology of name changes for an
organism.
[0127] Next, the persistent identifier is used to make a query 530
against the N4L-CI database 14. This query returns an additional
actionable code snippet and/or document fragment, typically written
in JavaScript, which identifies other digital resources or digital
records of non-digital resources in which the same name-like-string
or other name-like-strings which resolve to the same Exemplar
appear. The code snippet returned from the N4L-CI server 14 is
combined with the code snippet returned from the N4L server 12,
resulting in a single code snippet. This code snippet is then
associated 535 with the name-like-string in the document.
[0128] An example of how this N4L enhancement appears to the
End-User is shown in FIGS. 6-8. FIG. 6 shows a page from an online
resource that has not been enhanced. FIG. 7 shows the same page,
after the N4L enhancement code snippet has been added to the
resource, resulting in name-like-strings being made visually
distinct without interfering with the general presentation of the
document. FIG. 8 shows a dynamic pop-up menu activated by the
End-User's mouse-over of the N4L enhanced content. Each of the menu
items shown is actionable; the selection of any menu item results
in some specific action.
[0129] The shading of the items in the menu is significant. In this
example, the unshaded menu items are those which are offered for
free to the End-User. For example, by selecting the menu item
`Taxonomic tree`, a new browser window showing the relationship of
the selected organism to other organisms will be displayed.
Selecting the `Related materials` link will require that the
End-User pay a fee for accessing the information. The information
could include other articles, patents, etc., in which the same name
is used.
[0130] In another embodiment, an N4L persistent identifier,
particularly of the type Digital Object Identifier (DOI), resolves
to a DOI/Handle record containing multiple data fields, each of a
typed value, where the value type (e.g., URL, string) is known in
advance by software built on the N4L model. The application then
constructs graphical objects for the user to manipulate, for
example menu structures. Resolution to the values in the Handle
record allow the graphical objects to be populated, for example as
items in a menu.
[0131] For different classes of N4L object, e.g., Name, or
Exemplar, value types may vary. For example, one value type for
Name might be an enumerated value from `current` or `deprecated`. A
value type for Exemplar might be an integer representing the number
of names it has been classified with. The schema of value types for
a class of object is bound to its DOI application profile. By this
means, N4L resolution services are registered with the
International DOI Foundation and the Application Programming
Interface (API) for building resolved content into tools and
services is formally managed.
[0132] Two processes for enabling digital resources are described,
although other processes are also possible. The first is the
Workflow model for situations in which the original digital
resource is read-only and which results in the production of a new
enabled document, and the second is the Plug-in model for those
situations where the digital resource resides in memory and is
enabled without generating an entirely new document. Two approaches
for the Plug-in model are described.
[0133] FIG. 9 provides a flowchart 900 showing the process steps
for the Workflow model, where the flowchart 900 starts at 905 and
stops at 950. This is referred to as the Workflow model as this
processing is typically performed by Content Creators in an
off-line manner. In this case, the Content Creator has an original
document to be enabled by the presently described invention. The
output of the enablement is a new document, distinct from the
first, which is then possibly subjected to additional process steps
prior to publication. Note that the output is not necessarily web
content, i.e., HTML. Instead, the output can be XML, or any other
document format, which could then be subjected to additional
processing steps. The information embedded by the N4L program is
encapsulated in such a manner that it remains intact throughout the
successive processing steps.
[0134] The process for the Workflow model may be implemented as a
standalone program, as a client to a hosted service (such as a web
service), or via an extension to an existing application such as a
web browser or document editor (such as Microsoft Word or XML
Spy).
[0135] Following the start 905, the first step is to determine if
the content has already been N4L-enabled 910. The next step is to
identify all name-like-strings in the document (content), as
described below, and for each instance perform step 920 and either
step 925 or step 930 in a loop 915, until all instances have been
analyzed, when the loop ends 940. Any of a plurality of methods for
identifying name-like-strings may be employed, including, for
example, identification based on typographic convention, parsing
the document for strings formatted in a particular manner and/or
occurring in specific relations to other strings, and
identification based on a lookup table of known names. This process
can be performed by the program which implements the Workflow
process, by a N4L server, or by some combination of the two. The
process may search the graphical as well as the textual portion of
the content to find name-like-strings. The outcome of this process
is a determination of whether or not each name-like string is in
the database 920.
[0136] For those cases in which the name-like-string is found in
the database, the unique persistent identifier associated with said
name is returned (i.e. the name object identifier is retrieved from
the N4L server) 925 and is inserted into the document being
processed by associating it with the name-like-string. For example,
this can be accomplished with a XML metadata object that includes
the name-like-string and the persistent identifier. In addition,
the name and its inclusion in the digital resource is recorded in
the N4L-CI database 14.
[0137] For those cases in which the name-like-string is not found
in the database, the Content Creator will be made aware of this
fact (i.e. the name is flagged for analysis) 930, providing the
opportunity for one of several actions in response to this
situation.
[0138] A first action is to provide a list of names that are
similar to the name-like-string, i.e., it is assumed that there is
a typographical error. The list is presented to the Content
Provider, who is offered the opportunity to correct the error by
selecting the correct name from this list. The selected name is
recorded, the document is updated, and the information is stored in
the N4L-CI database 14.
[0139] A second action is for the Content Creator to ask the
document's author for clarification. The author's response would
dictate the necessary changes to the document, for example, if the
name is not validly published, it would be reformatted as a quoted
string.
[0140] A third action applies to a selected subset of publishers,
those whose publications are the ones in which new names are first
validly published. Within these publications, there will exist
sentences whose syntax indicates that an unrecognized
name-like-string is the definition of a new name. In these cases,
the tool would provide the publisher the means to notify the
curator of the N4L database of the new name, thereby providing a
means for the N4L database to be updated to include the new name.
This notification process may involve human intervention, may be
handled programmatically, or some combination of the two.
[0141] One of the outcomes from this process may be the
identification of errors in the original document. While human
readers are quite tolerant and forgiving of errors, machines are
not. Given the vast amount of information being produced, more and
more digital resources are being read and digested by machines. By
being able to identify and fix errors prior to publication of an
article, the value of the digital resource is greatly increased in
that it can now be accurately read by a machine reader. This will
lead to an improvement in the factors used to quantify the
quality/importance of an article, thereby increasing its value to
the publisher.
[0142] Once a document has been enabled 945 during production by
the association in the text of names with N4L identifiers, its
metadata is updated to record that this process has taken place,
for example in a META element.
[0143] FIG. 10 provides a flowchart 1000 showing the process steps
for the Plug-in model, the flowchart 1000 starting at 1005 and
stopping at 1050. This is referred to as the Plug-in model as this
processing is typically performed on a digital resource via an
add-on to an existing application. In this case, the process user
has an original document to be enabled by the presently described
invention. The processing is done in-memory, thus the output of the
enablement is a modification of the original document and is
typically not stored for future use.
[0144] The process for the Plug-in model, which runs on End-User
computer 16, is typically implemented as a browser extension,
although it can also be implemented as an add-on to any other
program which is used to view or edit digital resources. It could
also be a stand-alone program which can access network-accessible
digital resources.
[0145] The first step is to determine if the document has already
been N4L-enabled 1010. One manner in which this can be done is to
look for the existence of a META tag in the document header that
was produced prior to publication by the Workflow model process. In
this case, no further processing is required. Another is to
determine if the document has been previously processed via the
Plug-in model and stored in the N4L-CI database 14. In this case,
the N4L-enabled document may be returned from the N4L-CI database
14 to replace the current working version of the document.
[0146] The next step is to identify all name-like-strings in the
page, as described below, and for each instance to loop through one
or more of steps 1020, 1025, 1030, and 1035, until all instances
have been analyzed, when the loop ends at 1040. Any of a plurality
of methods for identifying name-like-strings may be employed,
including, for example, identification based on typographic
convention, parsing the document for strings formatted in a
particular manner and/or occurring in specific relations to other
strings, and identification based on a lookup table of known names.
This process can be performed by the program which implements
either the (1) Workflow or Plug-in model process, (2) by a N4L
server, or (3) by some combination of the two. The process may
search the graphical as well as the textual portion of the content
to find name-like-strings. In the case where some part of this
process is performed by the server, the plug-in can optionally send
the name-like-strings to the server individually; it can combine
all of the name-like-strings into a single document and send it to
the server; or it can send the entire original document to the
server. In the latter case, the server then has the ability to
store the document in a pre-processed form as described above. The
outcome of this process is a determination of whether or not each
name-like string is in the database 1020.
[0147] For those cases in which the name-like-string is found in
the database, the unique persistent identifier associated with said
name is returned (i.e. the name object identifier is looked up and
returned from the N4L server) 1025 and is inserted into the
document being processed by associating it with the
name-like-string (i.e. the document content is enabled) 1030. For
example, this can be accomplished with an XML metadata element that
includes the name-like-string and the persistent identifier. In
addition, the name and its inclusion in the digital resource is
recorded in the N4L-CI database 14.
[0148] For those cases in which the name-like-string is not found
in the database, one or more of several actions occur 1035. One
action is to identify recognized names that are similar to the
name-like-string and prepare an actionable code snippet and/or
document fragment with a list of possible names, which are then
made available to the End-User. While the name selected from the
list by the End-User may be recorded, this usage is not deemed to
be authoritative and is not stored in the N4L-CI database 14. A
second action is to store the unrecognized name-like-string, along
with the context in which it occurs, and to have an expert curator
determine if the string is in fact a name. An End-User's selection
of a name from the list of possible names may be used by the
curator to assist in determining the proper name to be associated
with the string. The determination made by the curator is deemed
authoritative, this authority is noted by the creation of a new
record within the N4L-CI database 14, and future viewings of the
web page will include enhancements associated with the proper
name.
[0149] One step in the N4L enablement process, as mentioned above,
is the identification of name-like-strings in the digital resource.
FIG. 11 provides a flowchart 1100, starting at 1105 and stopping at
1120, depicting the overall process for identifying the
name-like-strings contained within a digital resource. In one
embodiment, this process consists of two stages. First, an
algorithm extracts 1110 from the digital resource those strings
that are thought to be names, see FIG. 12. Second, the extracted
strings are then subjected to verification 1115, see FIG. 13.
[0150] As mentioned above, one method of identifying occurrences of
name-like-strings in a digital resource is to check each word in
the target resource against a database of known name-like-strings,
in a "brute-force" manner.
[0151] Other methods involve identifying name-like-strings based on
typographic, formatting, and/or stylistic conventions that are used
when referring to name-like-strings, FIG. 12. For example, in
formal taxonomic writing, name-like strings can be recognized based
on a series of typographic and stylistic conventions that serve as
visual cues to human readers that are prescribed by the various
Codes of Nomenclature. Biological names are generally Latinized,
which by convention results in these names appearing in italics at
all taxonomic ranks for Bacteria and Archaea and for genera (plural
of genus) and species for plants, fungi, yeasts, protists and
animals. In some instances, names may also have rank-specific
suffixes (e.g. -acea=family, -ineae=suborder, -ales=order,
-idea=subclass).
[0152] In formal taxonomic proposals that include a prologue and a
diagnosis (also referred to as a description or circumscription of
the taxonomic concept along with the name and its etymology), the
Latinized name appears in italics and is typically followed by an
authority string (the name of the author(s) proposing the original
taxonomic assertion, the date, and possibly the page of the
original published description. In instances where the name has
undergone subsequent revision, the authors of the latest revision
and accompanying cues regarding the nature of the revision are also
included in the authority string). The name-like string may also
contain information about the nomenclatural event and accompanying
taxonomic assertion by phrases such as "species novum", "combinatio
novum" in full or abbreviated form. Additional clues may gleaned
from the name based on the suffix (family--aceae; suborder--ineae;
order--ales; subclass--idea.
[0153] In normal usage, names appear in a less highly stylized
format and may exclude most of the supplementary information
regarding the authority for the name. By convention, names may also
appear in an abbreviated form (e.g., Bacillus subtillis=B.
subtillis). Formatting of the name in italics is, however,
regularly preserved to indicate that the name has standing in the
literature (and various databases). If the name appears in upright
text and is surrounded by quotation marks, it is deemed to be a
name without formal standing in the taxonomic literature.
[0154] Discovery of name-like strings in digital content can be
accomplished by examining digital content for the presence of
various tags or other embedded processor instructions that are used
to trigger a display device, a printer, or a digital typesetting
system to format a particular portion of the text in italics.
Name-like strings are then examined for additional known properties
including the text immediately following the string for the
identity of the naming authority, the presence of various modifiers
indicating taxonomic rank and name status. As only a portion of
strings appearing in italics are names, additional logic based on
the appropriate stylistic and typesetting conventions can be added
into the process of discovering other names and excluding incorrect
names. Name-like strings are then compared to a database of known
names with standing in the literature to determine whether or not a
particular name-like string has previously been encountered, and
the string is flagged for expert curation if appropriate.
[0155] FIG. 13 provides a flowchart 1300, starting at 1305 and
stopping at 1340, depicting one embodiment of the Name Extraction
process for biological names. In the case of biological names, the
names often appear italicized in text. For this reason, one
embodiment of the name extraction process is to identify and
extract all italicized strings within the digital resource (page)
1310. This is simply done by searching the resource for the tags
which indicate that the text located between the tags is
italicized.
[0156] Once the initial identification has been completed, a next
step of eliminating from future consideration known non-name
strings is performed 1315. For example, identification via
typographic convention may result in the finding of the string "et
al." One embodiment for performing this step is to filter the
preliminary list against a dictionary of italicized terms that
appear frequently but are not names. The filtering process will
remove many of these known non-names.
[0157] To each remaining item in the list of name-like-strings, a
unique identifier (such as a simple numeric index) is then created
and assigned for each phrase 1320. In one embodiment, these items
are compiled into a message (a new name verification request) 1325
that is sent asynchronously to the N4L server 1330 for
verification. The server then responds asynchronously with a list
of verified names with their associated DOI 1335.
[0158] FIG. 14 provides a flowchart 1400, starting at 1405 and
stopping at 1480, depicting one embodiment of the Name Verification
process. When a name verification request is received 1410, each
string in the list is examined to ensure it conforms to the
bacterial nomenclature format 1415, e.g., "Xyz abcde" would be
considered a possible full name and "Xyz. abcde" would be
considered a possible abbreviated name, so both are name-like
strings. In regular expression terms, this would be the equivalent
of "[A-Z][a-z]* [a-z]+" or "[A-Z][a-z]*. [a-z]+". The list of
potential names is then filtered against a second dictionary, to
exclude name-like strings that are known to be invalid 1415. Each
name-like string in the filtered list is examined and placed into
one of two lists: 1) a list of full names, and 2) a list of
abbreviated names, to separate abbreviated names from full names
1420.
[0159] A loop 1425 is performed for each full name, the loop ending
at 1450: Each name-like string in the list of full names is then
looked up in the N4L database 1430. If the name is not found in the
database, some additional action is taken, as described above, e.g.
the name is added to a queue for expert curation, 1445. If the name
is found in the database, the DOI is returned, and the name/DOI
pair is added to a list of verified names 1440.
[0160] After the list of full names has been processed, a loop 1455
is performed for each abbreviated name, the loop ending at 1470.
Each name-like string in the list of abbreviated names is matched
against the list of verified names to determine if each abbreviated
name matches a name in the verified name list 1460. For each name
that the abbreviated name matches in the list of verified names,
the abbreviated name and its matching name's DOI is added to the
verified name list 1465. After checking each abbreviated name, the
list of verified names and their associated DOI's are then returned
to the client 1475.
[0161] FIG. 15 provides a flowchart 1500, starting at 1505 and
stopping at 1550, depicting a high-level overview of the database
curation process. A loop is executed for each name-like string in
the curation queue 1510, the loop ending at 1540. Each name-like
string is examined by an expert curator 1515 and it is determined
whether the name-like string is valid 1520. If the name-like string
is determined to be invalid, it is added to the Exclusion
Dictionary 1525 and removed from the curation queue 1545. If the
name-like string is determined to be a valid name, the curator adds
the name and appropriate metadata to the database 1530, and assigns
a new DOI 1535. The name is then removed from the curation queue
1545.
[0162] One difficulty facing users of taxonomic data such as that
being operated on by the currently disclosed invention, including
for example researchers and data curators (e.g. persons who execute
the process described in FIG. 15), is understanding the taxonomic
and temporal relationships between the information objects. In the
case of biological nomenclature, these relationships arise due to
events such as literature events (e.g. the publication of a naming
Event), which result in the creation of an object, the creation of
a connection between objects, or a change in the status of an
object.
[0163] It is known that humans best understand concepts when they
are presented visually. Given the manner in which data are stored
in a database or are presented as lists, they are not easily
amenable to human intuitive understanding. Thus, for these types of
data, two coupled visualizations will assist the users'
understanding of the data.
[0164] An embodiment of the first visualization method, shown in
FIG. 16, is in the form of a timeline, with nomenclatural events
creating branches from the original name. This visualization does
not contain the taxonomic relationships. Instead, it is a chart of
objects showing their status and interconnections. The chart
presents each name along with all of the nomenclatural events, such
as effective publication date, valid publication date, corrections
and new combinations. Each of these events is asserted by a
literature reference and is stored in the N4L database. An
automatically-generated timeline chart such as that shown in FIG.
16 presents one or more objects using a visual indicator, which in
FIG. 16 is a horizontally-oriented bar. The visualization also
shows relationships among the objects and between the objects and
the timeline using dashed lines and arrows. Nevertheless, other
visual depictions of the temporal relationships of the objects can
also be employed. In addition, the terms `horizontal` and
`vertical` are arbitrary distinctions which used for convenience in
the present discussion, e.g. the axes and data shown in FIG. 16
could be reversed such that the timeline is vertically
oriented.
[0165] In this embodiment, the temporality of the events is shown
using the timeline along the horizontal (time-based) axis. The
temporal divisions (time points) may be evenly spaced or,
alternatively, may be compressed or expanded at various points to
provide more even spacing of dates and events or for other
considerations. The objects in this visualization can be vertically
arranged in order to group together like species or for other
considerations, or the vertical arrangement of the objects may be
arbitrary. Several vertical orderings of the objects, any of which
can be selected by the user when the diagram is being generated,
include alphabetical by name, chronological by date of first usage,
and minimization of crossing lines.
[0166] FIG. 17 presents the same visualization as FIG. 16 with the
addition of a second visual indicator such as a thick line, which
in one embodiment can be colored to more readily differentiate it
from the background. This visual indicator (line) indicates the
preferred name at any given time, e.g. the name preferred by
experts in the field. There can only be one preferred name per
exemplar at any given time. This line is conceptually similar to a
critical path through a Gantt chart. Other methods of identifying
the preferred name as a function of time can also be employed, for
example by altering the color or shading of the bars or by other
appropriate markings.
[0167] In one embodiment, the computer interface allows a user to
pick a point in time at which a taxonomic view is desired, as shown
in FIG. 18, by obtaining a selected date from the user, i.e.
obtaining from the user the location of a line crossing the
time-based axis. A taxonomy diagram is rendered from a single
vertical `slice` of this diagram (line 19-19'), representing a
snapshot of the nomenclature at a specific point in time. In
various embodiments, the user can select the point in time for
producing the taxonomy diagram using any of a number of mechanisms
to select a position on the timeline, e.g. by clicking with a mouse
or other pointer, sliding a pointer along the timeline, or typing
in a date, to name a few possibilities.
[0168] The taxonomy visualization, as shown in FIG. 19, presents a
snapshot at a given point in time of all related names and
illustrates some of the nomenclatural and taxonomic connections
between them. This figure, which was rendered from the underlying
N4L database, shows a snapshot of the nomenclature related to
Streptoverticiffium and Streptomyces from November 2009. By
changing the location of where the vertical line 19-19' crosses the
timeline axis in FIG. 18, a taxonomic visualization can be produced
for other time points.
[0169] With these two visualization methods, a user can better
understand the relationships between the information objects and
how these relationships change over time. One embodiment of a
method 2000 to create these visualizations, using the domain of
biological nomenclature as an example, is shown in FIG. 20.
[0170] The method 2000 begins at start 2005 with the user selecting
both a name object and a method of ordering (ordering basis) 2010.
The N4L database is searched and related name objects are retrieved
2015. Relatedness of name objects may be determined in various
ways, for example as described by Garrity et al. (US Pat. Appl.
Publ. No. 2005/0160059). An ordered, in-memory list is created of
all related name objects in the N4L database, which includes
various items of time-based publication information including, at a
minimum, the publication date of the name object, and the date and
type of other key events, such as corrections, change in status,
etc. Starting with the selected name object, and looping through
all name objects in the list 2020, time-based publication
information is extracted 2025 from the selected name object and all
related name objects identified in the N4L database. Using the
method of ordering (ordering basis) selected by the user, the name
objects are sorted into a list order, where the list order is used
to determine a relative arrangement of the name objects. Next, a
visual indicator such as a bar is created and placed onto the
drawing field 2030. The horizontal position of the bar is
determined by the publication date extracted in step 2025 and the
vertical position is based on the list order. Next, the bar is
shaded and arrows are added to indicate the timing and type of the
literature events 2035. This process continues until all names in
the list have been processed, at which point the loop ends 2040 and
the method stops 2045. The result is that an output image such as
those shown in FIGS. 16-18 is generated. In the examples shown in
FIGS. 16-18, the output images are graphs including the selected
name object and the related objects retrieved from the N4L
database. The graphs include a time-based axis, such that each of
the objects is represented on the graph by a first visual indicator
which depicts the time-based publication information extracted for
each respective name object. In various embodiments the output
image is presented to the user using an output device, for example
a computer screen or a printer.
[0171] FIG. 21 presents an embodiment of a method 2100 for creating
a taxonomy visualization such as that shown in FIG. 19, which is
generated in response to a user selecting a date defining a point
in time, e.g. by identifying the position of the vertical line, as
shown for example in FIG. 18. The vertical line in FIG. 18
intersects all objects in the system at the selected point in time,
allowing for the creation of a view containing the set of
intersected objects and their state at the specified time. This
embodiment describes a case wherein the name objects are biological
species, however, the name objects can be of any type previously
disclosed. Associated with each object is a taxonomic rank, for
example a subspecies, a species, a subgenus, a genus, a family, a
tribe, a suborder, an order, a division, a subclass, a class, a
subphylum, a phylum, a superphylum, a kingdom, or a domain. As
discussed below, the taxonomic rank may be used to form sublists of
the objects.
[0172] The method 2100 starts 2105 by obtaining a user-selected
name object and method of ordering 2110 and finding and retrieving
related objects from the N4L database 2115 to produce an ordered,
in-memory list of all related names. The next step is to loop
through the ordered, in-memory list of all related names 2120 to
determine which ones were in existence as of the date corresponding
to the vertical line, i.e. as of the selected date obtained from
the user 2125. Next, a number of sublists are formed or created
2130 based on, e.g., the taxonomic ranks associated with the
objects, after which this initial loop ends 2140. In the
illustrated example, each sublist contains the names of a single
genus, as can be seen in the two `columns` of names shown in FIG.
19. Starting with the sublist which contains the name originally
selected by the user, the names are placed into the drawing field
2145. Once the first list is completed, the remaining lists are
looped through 2150 and sorted according to taxonomic rank, as
above, and names from the sublists are placed onto the drawing
field 2155. To facilitate understanding, the subsequent lists may
be ordered to align related names in the subsequent sublists with
the names from the first sublist.
[0173] Next, visual indicators such as arrows are drawn to shown
synonym events and new combination events 2160. These
horizontally-drawn arrows correspond to the dashed vertical lines
shown in FIGS. 17 and 18. The arrows may be optionally labeled with
an identifier to indicate the associated literature event. After
the loop ends 2165, the method stops 2170.
[0174] Various known methods can be used to create the
visualization on a computer system in accordance with the
embodiments disclosed herein. One such method for transforming the
data into a graphical representation employs the Java programming
language and is presented within the context of a web browser,
although other methods can be used as well. One or more
visualization figures can be displayed in a single window or in
multiple windows.
[0175] Additional meta information and inferences about the data
can be determined using certain aspects of the current invention.
As described above, each time a name-like-string in a digital
resource is found in the N4L database 14, a record of that name's
(information object's) presence in the digital resource is noted in
the N4L-CI database 14. Other information, such as the number of
times the name (information object) occurs in a digital resource
and/or the number of times an information object is accessed, can
also be stored in the N4L-CI database 14. By acting on the data
stored in the N4L-CI database 14, additional information can be
extracted.
[0176] Tracking the number of times an information object is
accessed can be useful for certain end-users. To compile these
data, a computer process can count the number of requests being
made for all names (information objects). Each time a N4L-enabled
page is loaded, requests for N4L-enhancement for each identified
name in the digital resource are sent to the N4L Server 12, which
provides the means for performing the count. This count can include
the number of requests ever made for the name or the number of
requests which have been made in some period of time. This
information can be used to calculate the average number (and
standard deviation) of requests per time period for each name.
Should the average count value for a name (or a group of closely
related names per the network described above) change
significantly, this change can be noted. In one embodiment, "change
significantly" means a difference which is greater than
three-tenths of a standard deviation, although other levels of
change are also possible. Examples of responses to changes in name
request frequency include notifying one or more BRCs that there may
be additional demand for a specific item, notifying one of more
suppliers of reagents that there may be additional demand for a
specific item, and notifying law enforcement agencies of changing
patterns.
[0177] Examples of inferences which can be extracted from the data
stored in the N4L-CI database 14 include identifying publications
focused on similar topics, identifying authors working in similar
fields, and discovering heretofore unrecognized relationships
between objects. While there are numerous methods known in the art
for extracting such inferences, in one embodiment, the first step
is to create a graph which links all of the digital resources
stored within the N4L-CI database 14 to each other using a distance
metric. This is done by calculating a distance metric for each pair
of digital resources in the N4L-CI database 14. The distance metric
is a numerical value calculated by a mathematical formula which
determines the proximity, or relatedness, of the digital resources
to each other based on the number and strength of shared concepts.
A distance metric may produce either smaller or larger values to
indicate closer proximity, i.e., that the two digital resources are
more closely related to each other. Graph analysis techniques,
which use either smaller or larger values to indicate closer
proximity, are known in the art.
[0178] The following are examples of distance metrics, but this
method is not limited by these embodiments.
[0179] In one class of metrics, so-called symmetric metrics, the
distance from digital resource 1 to digital resource 2 is the same
as the distance from digital resource 2 to digital resource 1. Such
a metric can be calculated by counting the number of names
(information objects) shared by the two digital resources divided
by the total number of unique names that appear in both digital
resources. Given S1, the set of all names in digital resource 1,
and S2, the set of all names in digital resource 2, then an example
distance metric is given by:
d = S 1 .times. IS 2 S 1 .times. U .times. S 2 ##EQU00001##
[0180] An example of the use of this distance metric is graphically
depicted in FIG. 22. The N4L CI 14 database is comprised of six
digital resources, each with some number of names 2205 (FIG. 22A).
Using the above algorithm, the distance between each of the digital
resources is calculated and displayed in tabular form 2210 (FIG.
22B). For purposes of visualization, the tabulated distances can be
represented as a graph, in which the line between nodes shows the
strength of the relationship between the digital resources, 2215
(FIG. 22C).
[0181] Another possible symmetric metric is the summation of the
number of times a name appears in both digital resources divided by
the total number of name instances in both digital objects. Another
possible distance metric is based on the mutual occurrence of
uncommon names, where "uncommon" is defined in one embodiment as
"occurring in less than 1% of all digital resources in the N4L-CI
database".
[0182] In another class of metrics, so-called asymmetric metrics,
the distance from digital resource 1 to digital resource 2 is
potentially different from the distance from digital resource 2 to
digital resource 1. Such a metric can be calculated by counting of
the number of names (information objects) shared by the two digital
resources. Then, the distance from digital resource 1 to digital
resource 2 is given by the number of shared names divided by the
total number of unique names in digital resource 1; and the
distance from digital resource 2 to digital resource 1 is given by
the number of shared names divided by the total number of unique
names in digital resource 2:
d 1 .fwdarw. 2 = S 2 S 1 .times. I S 1 , d 2 .fwdarw. 1 = S 1 S 2
.times. I S 2 ##EQU00002##
[0183] Once the graph is built, any of a variety of algorithms
known to those of ordinary skill, can be used to analyze it. For
example, a clustering algorithm, such as that by Zahn (Zahn, C. T.
(1971) "Graph-theoretical methods for detecting and describing
Gestalt clusters," IEEE Transactions on Computers, C 20, 68-86),
can be used to identify groups of articles which address similar
subject matter. For the example shown in FIG. 22, such an algorithm
would identify two distinct clusters of digital resources. This
functionality could be monetized by offering to provide to users
references to these other digital resources, thereby saving them
from having to review them. Using the Practitioner field stored in
an information object, a clustering algorithm could also identify
researchers who are working in similar fields (i.e. the `discovery
of research networks`). This information is of value for
researchers (to identify potential collaborators and/or
competitors), employment agencies, and other individuals or
entities. These are but two illustrative examples of how the graph
can be used to extract metadata from the system.
[0184] In various embodiments, the present invention is implemented
on one or more computers 10, 12, 14, 16, 32, and 38, see FIGS. 1,
2, and 4. In those embodiments in which more than one computer is
used, the computers may be in operative communication with one
another through a variety of wired or wireless mechanisms, or
through physical transfer of computer-readable media among the
computers. The computers may be in communication via a local-area
network (LAN) and/or via the Internet or other large-scale computer
network. As shown in FIG. 23, each computer 120 may have an input
device 140, an output device 160, a storage medium 180, and a
processor 200. Possible input devices 140 include a keyboard, a
computer mouse, a touch screen, and the like. Output devices 160
include a cathode-ray tube (CRT) computer monitor, a liquid-crystal
display (LCD) computer monitor, printer, and the like. Storage
media 180 include various types of memory such as a hard disk, RAM,
flash memory, and other magnetic, optical, physical, or electronic
memory devices, whether fixed or removable. The processor 200 is
any computer processor capable of performing calculations and
directing other functions for performing input, output,
calculation, and display of data in the disclosed invention.
Various embodiments of the invention are implemented as a set of
instructions and data that are stored on the storage media 180 of
one or more computers. In various embodiments, one or more steps of
the claimed methods may be executed using the processor 200. In
various embodiments, the invention may include a computer program
product including a computer usable medium (e.g. storage media 180
as described above) having a computer readable program code
embodied thereon, where the computer readable program code is
adapted to be executed (e.g. using a processor 200) to implement
methods of the invention.
[0185] In various embodiments, the present invention can be used in
conjunction with a database such as that described by Garrity et
al. (US Pat. Appl. Publ. No. 2005/0160059).
[0186] While certain representative embodiments and details have
been shown for purposes of illustrating the invention, it will be
apparent to those skilled in the art that various changes may be
made without departing from the scope of the disclosure, which is
further described in the following appended claims.
* * * * *