U.S. patent application number 14/493256 was filed with the patent office on 2015-03-19 for system and method for collaborative knowledge structure creation and management.
The applicant listed for this patent is Philip Marshall. Invention is credited to Philip Marshall.
Application Number | 20150081689 14/493256 |
Document ID | / |
Family ID | 39102598 |
Filed Date | 2015-03-19 |
United States Patent
Application |
20150081689 |
Kind Code |
A1 |
Marshall; Philip |
March 19, 2015 |
SYSTEM AND METHOD FOR COLLABORATIVE KNOWLEDGE STRUCTURE CREATION
AND MANAGEMENT
Abstract
A collaborative, semantic-based knowledge structure creation and
management system and method are provided. The system and method
may include knowledge structures which include concept based maps
which may be linked according to topic, relationship types,
conceptual equivalence, and temporally. The knowledge structures
may be managed by users, wherein users may perform at least one of
adding a new concept, creating a new relationship, confirming an
existing relationship, and breaking an existing relationship.
Inventors: |
Marshall; Philip; (Portland,
OR) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Marshall; Philip |
Portland |
OR |
US |
|
|
Family ID: |
39102598 |
Appl. No.: |
14/493256 |
Filed: |
September 22, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11827563 |
Jul 11, 2007 |
8843475 |
|
|
14493256 |
|
|
|
|
Current U.S.
Class: |
707/725 |
Current CPC
Class: |
G06F 16/367 20190101;
G06N 5/02 20130101; G06F 16/2477 20190101; G06F 16/24575 20190101;
G06F 16/242 20190101; G06F 16/94 20190101 |
Class at
Publication: |
707/725 |
International
Class: |
G06F 17/30 20060101
G06F017/30 |
Claims
1. A method for linking a plurality of data concepts stored on a
network, the method comprising: receiving temporal information for
a first data concept; receiving temporal information for a second
data concept; temporally linking the first data concept and the
second data concept; graphically displaying a timeline showing the
relationship of the first data concept and the second data concept;
and receiving user input affirming the relationship of the first
data concept and the second data concept.
2. The method of claim 1, wherein further receiving user input
challenging the relationship of the first data concept and the
second data concept and displaying a timeline element indicating a
user challenge to the relationship.
3. The method of claim 1, further comprising receiving a search
query for one of the first data concept and the second data concept
and displaying the timeline.
4. The method of claim 1, wherein further receiving user input
includes adding further temporal information to one of the first
data concept and the second data concept, and displaying the
further temporal information on the timeline.
5. A software program, recorded on a computer-readable medium
having computer-readable instructions, which when executed, caries
out the method comprising: displaying a collaborative timeline
comprising a plurality of user-proposed, temporally-related
concepts; receiving an affirmation or a challenge to one or more of
the user-proposed, temporally-related concepts; and updating the
timeline according to the affirmation or the challenge.
6. The program of claim 5, the method further comprising receiving
additional user-proposed, temporally-related concepts.
7. The program of claim 5, the method further comprising updating
the timeline with the additional user-proposed, temporally-related
concepts.
8. The program of claim 5, the method further displaying a subset
of the timeline in accordance with a user request.
9. A method of generating a timeline display, the method
comprising: receiving a first temporal concept in the form of a
search query from a user; setting a granularity of a timeline to
search for events over a select period of time; retrieving a second
temporal concept pertaining to the search query and the granularity
of the timeline; retrieving a plurality of sub-concepts linked to
the second temporal concept; where the sub-concepts are returned
based on the granularity of the timeline and the strength of the
link between the second concept and the sub-concepts, the strength
of the link established on affirmation of a proposed collaborative
topic-to-topic relationship linkage; and displaying the second
concept and the sub-concepts on the graphical timeline, where the
sub concepts are displayed based on the granularity of the timeline
and the strength of the link between the concept and the sub
concepts.
10. The method of claim 9, further comprising prompting the user to
vote to affirm or decline the second concept.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] This application is a continuation of U.S. patent
application Ser. No. 11/827,563, filed on Jul. 11, 2007, and titled
"SYSTEM AND METHOD FOR COLLABORATIVE KNOWLEDGE STRUCTURE CREATION
AND MANAGEMENT" the entire disclosure of which is incorporated by
reference for all purposes.
BACKGROUND AND SUMMARY
[0002] Recently, a vast amount of information about a multitude of
different types of topics has become accessible through searches
using search engines on networks. Some search engines have been
configured to provide fast responses to search requests while
retrieving highly clustered and indexed sets of links. The links
may be prioritized by the number and kind of websites that link to
them. These types of search engines may perform different website
identification and organization, for example, web crawling, sorting
and normalizing website links, and prioritizing website links.
[0003] In some situations, the search engines may provide results
which are either too narrow, providing only a small number of
websites with meaningful content, or too broad, providing a
quantity of websites so large that it may become cumbersome trying
to search every website for meaningful content. Furthermore, search
engines generally do not provide orientation around a topic, and
even searches on synonymous words for the same topic can produce
drastically different results. And since search results are
generally not oriented by topic, the ability to navigate between
related topics may not be available.
[0004] Some emerging search engine technologies, such as Google
Co-op, are, however, beginning to enable humans to index websites
with certain topic-specific metatags that then facilitate
topic-to-topic navigation. Related is the way in which systems,
such as inventory systems, organize data. While historically these
systems have relied upon proprietary database schemas to store
data, enabling retrieval through proprietary search queries,
emerging information retrieval technologies are enabling faster
retrieval of this information and even more intuitive display
through the use of synonyms and topic-to-topic relationships and
navigation.
[0005] Thus, due to the emergence of search technologies that
utilize synonyms and topic-to-topic relationships to facilitate
better information retrieval and display, there is a need for a
system and method that can facilitate the creation of definitive,
trusted "knowledge structures" of synonyms between terms for a
given topic, and of relationships between topics. Such a system and
method used to formalize a knowledge structure might be used by a
person or groups of persons considered trusted to establish such
structures, or might be used in a more open, collaborative way,
enabling a large number of users of a network to formalize the
knowledge structures through more mathematical formalization
processes.
[0006] For example, knowledge structures may be created by a group
of networked individuals using the knowledge structure creation and
management system, and the resulting knowledge structures may then
be used to facilitate information retrieval, such as for use in a
search system or in an inventory management system.
[0007] In one aspect of the system, concepts and events may be
linked temporally via the knowledge structures and searching the
knowledge structures may result in a timeline of events that may be
displayed to a user. The timeline may include events that relate to
a concept temporally and according to other relationship types. The
temporal linking of concepts may facilitate intuitive searching
between concepts that may be linked through time.
[0008] Further, in another aspect of the system, users may
continuously manage the system by revising knowledge structures
within the system. Knowledge structures may be updated by creating
or changing links between elements, adding new elements, removing
links, and removing elements. Moreover, the management process may
be conducted through the collaborative efforts of users of the
system.
BRIEF DESCRIPTION OF DRAWINGS
[0009] The disclosure is illustrated by way of example and not by
way of limitation in the figures of the accompanying drawings, in
which the like references indicate similar elements and in
which:
[0010] FIG. 1 is a schematic representation of a network in which
an exemplary embodiment of the knowledge structure creation and
management system of the present disclosure may be implemented.
[0011] FIG. 2 is a schematic diagram of an exemplary embodiment of
a knowledge structure creation and management system.
[0012] FIGS. 3A-3C are schematic representations of partial
semantic maps of a knowledge structure creation and management
system.
[0013] FIG. 4 is a schematic representation of concepts linked
together based on user characterizations.
[0014] FIG. 5 is a schematic representation of a network in which
an exemplary embodiment of the knowledge structure creation and
management system includes different types of users.
[0015] FIG. 6 shows example search processes performed by a trusted
user of the knowledge structure creation and management system of
FIG. 5.
[0016] FIG. 7 shows example search processes performed by a general
user of the knowledge structure creation and management system of
FIG. 5.
[0017] FIG. 8 shows an example flow diagram for creating and
managing knowledge structures in a general user model.
[0018] FIG. 9 shows an example flow diagram for creating and
managing knowledge structures in a trusted user model.
[0019] FIG. 10 shows an example conceptual search graphic user
interface depicting concepts linked according to knowledge
structures of the knowledge structure creation and management
system.
[0020] FIG. 11 shows an example graphic user interface depicting
concepts linked temporally on a timeline according to knowledge
structures of the knowledge structure creation and management
system.
DETAILED DESCRIPTION
[0021] Systems and methods to create collaborative semantic-based
knowledge structures are described below. Although the systems and
methods of searching may be described in one example in regards to
searching websites on a network, it should be appreciated that such
systems and methods may be used in a variety of environments,
including, but not limited to, logistical operations, marketing and
advertising, law enforcement, etc. and used as part of general
information retrieval or within specialized areas such as law
enforcement, auto parts, or insurance.
[0022] FIG. 1 illustrates a schematic diagram of a network in which
an exemplary embodiment of a collaborative semantic-based knowledge
structure creation and management system of the present disclosure
may be implemented. The network (generally referenced as 100)
includes a plurality of user computing devices (alternatively
referred to as user computers) 102. The user computing devices 102
may be any sort of device capable of connecting to and
communicating with the network 100, for example, a desktop
computer, a laptop computer, a personal digital assistant, a smart
phone, etc. Furthermore, the network 100 may include one or more
repositories 104 which may be in communication with the user
computers 102.
[0023] It should be appreciated that an exemplary embodiment of the
knowledge structure creation and management system, which may be
used by a search system in one example, may be implemented on a
variety of different networks, including public networks, private
networks, shared networks, etc. For example, the knowledge
structure creation and management system may be configured for use
with searching for information stored on a public network system
such as the World Wide Web or Internet, a private system such as an
intranet network, or an individual computer. Furthermore, the
knowledge structure creation and management system may be
configured for use with searching for information on newsgroups,
open directories, miscellaneous databases, etc.
[0024] In the exemplary knowledge structure creation and management
system, users may perform search queries to repositories on the
network in order to retrieve information stored in the
repositories. Stored information may relate to any sort of data
that may be conceptually organized, including, but not limited to,
website identifier information, personal information, geographical
information, etc.
[0025] Information stored in the repositories may be organized such
that, for example, terms, phrases, and other data strings may be
categorized into concepts (alternatively referred to as topics).
These concepts may be semantically linked together through
taxonomic relationships. For example, a concept may be linked to a
multitude of other concepts based on the type of relationships that
exist between the concepts. In some embodiments, the concepts may
be linked temporally, and/or linguistically, such as an
identification of synonymic concepts. Accordingly, the linked
concepts may produce an array of semantic maps, which may be
traversed by search queries.
[0026] Furthermore, the semantic maps may be used to define
structured filters (alternatively referred to as knowledge
structures) that may aid in information retrieval. For example, the
semantic organization of concepts may enable concept-based
browsing, and cross-referencing to types or subtypes of concepts.
As another example, a search query, using the structured filter
approach may return results pertaining to a particular topic and,
in addition, may return people, places, and products relating to
the topic. As another example, a search query may return temporally
linked concepts organized in chronological order. In an exemplary
embodiment of the knowledge structure creation and management
system, concepts and related information may be more accessible to
a user due to the filtering achieved by the knowledge structures
that may correlate concepts and terms according to the preferences
of user of the knowledge structure creation and management
system.
[0027] In some embodiments of the knowledge structure creation and
management system, a knowledge structure may be created by one or
more users who may link different elements to a particular topic.
As an example, a particular knowledge structure may start as a
single topic linked to several related topics. Further, users of
the system may collaboratively expand the knowledge structure
during use of the system by adding additional elements and/or
links. As a knowledge structure is developed based on user
interaction, a map of conceptually linked topics may be created,
wherein users of the system may traverse the concept-based map to
filter terms and concepts in order to retrieve information while
browsing a particular topic.
[0028] An exemplary knowledge structure system further may enable
users to continuously manage various knowledge structures in the
system. For example, in some embodiments, the system may enable
users to change or remove established links between synonyms
(intra-concept), between topics (inter-concept), and/or temporally.
Additionally, the system may enable users to remove particular
elements if desired.
[0029] Moreover, in some embodiments, the system may enable users
to completely remove entire knowledge structures from the system if
desired. Furthermore, various knowledge structure system
configurations and use examples are described in more detail
below.
[0030] Now referring to FIG. 2, a high-level, schematic
representation of an exemplary knowledge structure 200 of the
present disclosure, used in an exemplary search system, is shown.
Typically a knowledge structure can be used in a search for a
desired topic. A user may create an input 202, for example, a text
string such as a term, phrase, or request. The input 202 may be
converted into a query of the repository. The query may be filtered
through the knowledge structure 200 in order to retrieve a
conceptual match to the query. Furthermore, the resulting
conceptual match may be produced as an output 204, along with other
topics linked to it within the knowledge structure.
[0031] It should be appreciated that in some embodiments, the input
202 may include visual elements, graphics, audio excerpts, etc.
Accordingly, the system may produce conceptual matches represented
by visual elements, graphics, audio excerpts, etc.
[0032] With use of the knowledge structure creation and management
system, the system may enable an input query to be mapped to
concepts within a knowledge structure. The input query may traverse
linked elements throughout a semantic map while filtering desired
conceptual results. For example, the structured filter may retrieve
information which conceptually relates to the input. In the
illustrated embodiment, the structured filter 200 retrieves two
concepts (Concept 1 and Concept 5) determined to be conceptually
related to the input query. The knowledge structure creation and
management system may output the retrieved concepts, and
additionally, the output may include other concepts (Concept 2-4
linked to Concept 1; Concept 6-8 linked to Concept 5) including
multiple sub-level concepts which may be linked to the retrieved
concepts. For example, a concept retrieved by a search may be
related to multiple concepts on a first sub-level, and further may
have relationships to other concepts on second, third, fourth, etc.
sub-levels. In some embodiments, the sub-levels of additional
concepts may be organized in the output according to relationship
type.
[0033] FIG. 3A shows another example of a schematic representation
of a semantic map of concepts linked according to types of
conceptual relationships illustrating use of a knowledge structure.
In the illustrated embodiment, for example, a structured filter may
retrieve a concept (Concept 1) which may be linked to two
additional concepts (Concept 2 and Concept 3) through a first type
of relationship (Relationship 1) and two more additional concepts
(Concept 4 and Concept 5) through a second type of relationship
(Relationship 2). Some example relationship types may include, but
are not limited to, type of, is part of, is manufactured by,
transports, works for, was before, occurred before, occurred after,
etc.
[0034] The above described use of the exemplary knowledge structure
creation and management system may include semantic maps of
concepts organized according to conceptual relationship type.
However, it should be appreciated that in some embodiments of the
knowledge structure creation and management system, a structured
filter may be organized by conceptual equivalence (i.e. synonym)
and/or temporally.
[0035] Moreover, a user or group of users may create knowledge
structures comprising of synonyms for a given topic and/or may
create the knowledge structures comprising of relationships between
topics. Examples of such creation and/or management of knowledge
structures are described in more detail below.
[0036] Referring to FIG. 3B, a nonlimiting example of a partial
semantic map organized by conceptual equivalence is shown. In the
illustrated embodiment, an input query may be mapped to a concept
(Concept 1) which is linked to four character strings (Character
Strings 1-4) based on conceptual equivalence. In other words, the
system may map an input query to a concept within a knowledge
structure, and the system may then retrieve and output the concept
as well as the character strings that are synonymous with the
concept. The knowledge structure creation and management systems
determination of conceptual equivalence between two character
strings will be discussed in more detail below.
[0037] FIG. 3C shows another example of a partial semantic map
organized by conceptual equivalence. As shown, a concept may be
linked to four character strings. Three of the character strings
may be classified as synonyms and the fourth character string may
be classified as a routing string. A routing string may not
represent the exact meaning of a concept but may often be employed
by users when trying to identify a concept, for example a euphemism
or colloquialism.
[0038] Additionally, in some embodiments, the concept may be linked
to a preference string which may be a character string with is
determined to have a priority level of audience preference when
representing the concept. In some cases, the preference string may
be established through use of the structured filter and output may
be continuously refined according to user preference. In one
particular example, a preference string may be established based on
user clicks corresponding to a retrieved concept. Furthermore, a
concept may have multiple levels and sub levels of prioritized
strings. In some embodiments, audience preference may include the
collaborative actions and intentions of multiple users of the
knowledge structure creation and management system.
[0039] In some embodiments, a knowledge structure may be configured
such that text strings may be mapped into concepts or topics, such
as, for example synonyms, preferred strings, routing strings, and
temporal strings. Knowledge structures with such a configuration
may be characterized as having intra-topic mappings. Further in
some embodiments, a knowledge structure may be configured such that
topics may be mapped to multiple other topics through various types
of relationships. Knowledge structures with such a configuration
may be characterized as having inter-topic mappings.
[0040] In some embodiments of the knowledge structure creation and
management system, users may dynamically manage knowledge
structures of the system by using various management tools to
modify a particular knowledge structure. For example, the system
may be configured with an online tool. The tool may be used to
query a knowledge structure to retrieve information.
[0041] In some embodiments, the tool further may be configured to
enable a user to directly confirm or break a relationship between
two concepts within the knowledge structure. For example, a user
may perform a search for a character string using the tool. The
search may return several conceptually equivalent, synonymous,
and/or temporal character strings. The knowledge structure creation
and management system may be configured to enable the user to
determine if the returned character strings, do in fact, have a
legitimate relationship. The user may have the ability to confirm
or break the link between the searched character string and the
retrieved concepts. In an exemplary embodiment of the knowledge
structure creation and management system, users may determine which
concepts may be linked together and through which type of
relationship they may be linked. Furthermore, groups of users may
collectively manage the knowledge structure system by making
determinations about included elements and links.
[0042] Additionally, in some embodiments the system may include
tools to add or remove various knowledge structures. Thus,
collaboratively, users of the knowledge structure creation and
management system may contribute to the management of knowledge
structures within the system.
[0043] Furthermore, users of the knowledge structure creation and
management system may continuously expand or contract the semantic
mapping of concepts in a structured filter. For example, if a user
is searching for a concept and nothing is retrieved by the
knowledge structure creation and management system, the system may
prompt a user to add a new concept. In turn, the user may propose a
new concept to be added to the semantic map. Additionally, a user
may propose that new links be created between concepts if they do
not exist in a particular knowledge structure. Conversely, if a
user finds that two concepts are linked together and determines
that the link is not legitimate, the user may propose to break the
link between the concepts. User proposals may be formalized in the
knowledge structures according to different establishment or
breaking criteria described herein.
[0044] Referring now to FIG. 4, a schematic representation of a
partial semantic map is shown with user defined confirmations or
breaks of links between concepts. In the illustrated embodiment,
concepts 400 are shown linked to one another. For each link a user
may propose to confirm the link or break the link. In other words,
users may cast votes to create or break relationships between
concepts. As shown, the user proposed confirmations of a link are
depicted as squares 402, while user proposed breaks are depicted as
crosses 404. Furthermore, users may propose new concepts 406 and
links, which are shown in dashed lines. In one example, three
proposed linkages (squares) may be required to confirm the link and
three proposed breaks (crosses) may be required to break the link
in an open model, whereas only a single proposed linkage may be
required to confirm the link and one proposed break may be required
to break the link in a trusted model.
[0045] As described above, in some embodiments, the knowledge
structure creation and management system may determine a preference
string to represent a concept as a conceptual equivalent. A
preference string may be a string with a high level of priority
based on audience preference when representing a concept. In other
words, the priority level of a preference string may be determined
by the amount of user proposals for a particular concept.
Specifically, for example, if a concept is linked to multiple
synonyms then the synonym with the most user confirmations may be
determined to be the preference string for the concept.
[0046] In some embodiments the knowledge structure creation and
management system may include different types of users. As shown in
FIG. 5, for example, a general user 502 and a trusted user 504 may
utilize an exemplary knowledge structure creation and management
system and may search a knowledge structure creation and management
system repository 506 over a network 508. Within the system, the
trusted user may be able to create changes in a knowledge
structure, for example, create or break links between concepts and
add new concepts directly without having to collaborate with others
to formalize the change. Thus, in some cases, the establishment
criteria for the trusted user may be a proposal for change of the
knowledge structure. In some embodiments, the general user of the
knowledge structure creation and management system may be permitted
to support proposed linkages or breakages in links between
concepts, and may proposed new concepts and/or links. However, the
general user may only contribute to a model of formalization for a
concept because the system may require that multiple general users
propose a linkage or break until a mathematical threshold of like
proposals is met, thereby formalizing the change or addition. Thus,
in some cases, the establishment criteria for the general user may
be affirmation of a proposal from one or more users. In this way,
general users may collaborate to create and manage one or more
knowledge structures. Moreover, changes in the knowledge structure
may be formalized according to establishment criteria including for
example a plurality of similar proposal that may exceed a
mathematical threshold.
[0047] Another exemplary embodiment of the knowledge structure
creation and management system may employ a model of formalizing
inter-concept mappings that may be more automated. Also, an
exemplary embodiment may include less deliberate manual modeling by
users to manage knowledge structures in the system. In such a
model, for example, users of a search system may navigate
particular paths to certain content following their search, such as
navigating to sub-types of a topic. In one configuration, the
system model may utilize the frequency of navigation to those
sub-types to formalize the relationship of those topics as
sub-types of the originally searched-upon topic.
[0048] For example, users of a search system may search and
retrieve information about a famous singer, and then navigate
through a link called "songs" to that singer's musical works. If a
defined mathematical threshold is met based on the number of users
navigating such links, then the knowledge structure will create a
relationship between the "songs" and the "singer" through a
relationship called "songs of" or "performed by".
[0049] It should be appreciated that in some embodiments, an
automated model may be used in combination with a mathematical
threshold model (discussed above) or a trusted user model
(discussed below) to formalize elements in a knowledge structure
creation and management system.
[0050] In some embodiments of the knowledge structure creation and
management system, a group of trusted users may act as managers of
the knowledge structure creation and management system and may have
additional organizational and administrative privileges. A user may
be deemed a trusted user for a variety of reasons, such as being an
expert on a particular topic, working as part of the knowledge
structure creation and management system operational management
team, etc. In the trusted user model, a trusted user may modify a
knowledge structure, for example add new concepts and create or
break links between concepts without collaboration of other users
to formalize the modification. The trusted user model may assume
that the trusted user may have expertise in a particular topic and
therefore may be granted additional management privileges.
[0051] In some embodiments of the knowledge structure creation and
management system, both the general user model and the trusted user
model may be employed. In such a system, general users may propose
various additions and/or modifications to the knowledge structures
that require collaborative support of other users to formalize the
modification. Whereas, trusted users of the system may propose
various additions and/or modifications to the knowledge structures
and the modifications may be formalized without further support of
other users. In other words, trusted users may make direct changes
to the knowledge structure creation and management system; while
general users may be limited to making proposals to change the
knowledge structure creation and management system.
[0052] In some embodiments, when a general user proposes a change
to the knowledge structure creation and management system, the
general user may cast a vote to make the change. The votes
(proposed changes) of general users may be tallied for each link or
concept in the semantic map. Once a vote tally reaches a particular
mathematical threshold, a relationship between concepts may be
confirmed or broken, or a new concept may be added to the semantic
map. In some embodiments, different mathematical thresholds may
exist for different aspects of the knowledge structure creation and
management system. It should be appreciated that in some
embodiments, proposed changes made by general users may be accepted
according to a statistically significant formalization, or other
determining factors.
[0053] Now referring to FIG. 6, four examples of a trusted user's
process of creating and/or managing (e.g. modifying) a knowledge
structure is described. In a first input 602, a trusted user may
enter a term in a search query, for example, of the knowledge
structure creation and management system. The trusted user may
retrieve a concept which is linked to a second concept. Further,
the trusted user may choose to break the relationship between the
first and second concept. Thus, since the user is a trusted user
the break in relationship is formalized in the knowledge structure
creation and management system.
[0054] In a second input 604, the trusted user may enter a first
term in a search query of the knowledge structure creation and
management system. The trusted user may find a first concept
correlating with the first search term. Further, the trusted user
may enter a second search term into a search query of the knowledge
structure creation and management system and may retrieve a second
concept which correlates to the second search term. The trusted
user may link the first concept to the second concept through a
relationship. Thus, since the user is a trusted user the
relationship between concepts is formalized in the knowledge
structure creation and management system.
[0055] In a third input 606, the trusted user may enter a term into
a search query of the knowledge structure creation and management
system. The search query does not return a concept. Accordingly,
the system may prompt the trusted user to add a concept. The
trusted user may add a concept and links to the concept to the
semantic map. Thus, since the user is a trusted user the semantic
map of the knowledge structure creation and management system is
expanded to include the new concept and associated links.
[0056] In a fourth input 608, the trusted user may enter a first
term into a search query of the knowledge structure creation and
management system. The first term may not be found in the knowledge
structure creation and management system. Accordingly, the system
may prompt the trusted user to enter a new search query (although
is will be appreciated that the system may optionally or
additionally prompt the trusted user to add a new concept). The
trusted user may enter a second term into a search query of the
knowledge structure creation and management system. The trusted
user may find a concept related to the second search term. Further,
the trusted user may add a link between the first search term and
the second search term as well as add a "synonym" type link between
the first search term and the concept. Thus, the synonym is linked
to the concept in the knowledge structure creation and management
system.
[0057] Note that such management of a knowledge structure is not
limited to information gained through searching, but rather these
are examples of how a user may identify concepts that may be
linked, or not linked, in the knowledge structure, expand the
knowledge structure, generate synonyms, etc.
[0058] Now referring to FIG. 7, four examples of a general user's
search process is described. In a first search 702, a general user
may enter a term in a search query of the knowledge structure
creation and management system. The general user may find a first
concept which is linked to a second concept. The general user may
propose to break the relationship between the first concept and the
second concept. The proposal may be added to the tally of proposals
by other general users. When the tally overcomes a mathematical
threshold the break in relationship may be confirmed in the
knowledge structure creation and management system.
[0059] In a second search 704, the general user may enter a first
term in a search query of the knowledge structure creation and
management system. The general user may find a first concept
correlating with the first search term. The general user may enter
a second search term into a search query of the knowledge structure
creation and management system and may retrieve a second concept
which correlates to the second search term. The general user may
propose to link the first concept to the second concept through a
relationship. The proposal may be added to the tally of proposals
by other general users. When the tally overcomes a mathematical
threshold the relationship between concepts may be confirmed in the
knowledge structure creation and management system.
[0060] In a third search 706, the general user may enter a term
into a search query of the knowledge structure creation and
management system. The search query may not return a concept. The
general user may propose the addition of a new concept and a new
link between the concept and the search query. The proposal may be
added to the tally of proposals by other general users. When the
tally overcomes a mathematical threshold the semantic map of the
knowledge structure creation and management system may be expanded
to include the new concept and associated link.
[0061] In a fourth search 708, the general user may enter a first
term into a search query of the knowledge structure creation and
management system. The first term may not be found in the knowledge
structure creation and management system. The general user may
enter a second term into a search query of the knowledge structure
creation and management system. The general user may find a concept
related to the second search term. The general user may propose to
add the first term as a synonym to the concept. The proposal may be
added to the tally of proposals by other general users. When the
tally overcomes a mathematical threshold the synonym may be linked
to the concept as a synonym in the knowledge structure creation and
management system.
[0062] In some embodiments, the collaborative nature of the
knowledge structure creation and management system may facilitate
the structured filters to be validated, scalable, and maintainable
by the general users and the trusted users. Furthermore, the
semantic nature of the knowledge structure creation and management
system may allow the use of synonyms to drive topic or
concept-based searching. Likewise, the temporal nature of some
systems may enable topic or concept based searching. Also, the
semantic nature of the knowledge structure creation and management
system may allow the use of relationships between concepts to drive
topic or concept-based searching. An exemplary embodiment of the
knowledge structure creation and management system further may be
used to refine attributes for a character string to aid in
searching, including but not limited to language. Some embodiments
of the knowledge structure creation and management system may be
used to replace cataloging as a reliable method of structuring
topics.
[0063] The exemplary embodiment of the knowledge structure creation
and management system described above may be used in a variety of
different applications relating to searching on computer systems
over a network, for example, the World Wide Web, a private
intranet, databases etc.
[0064] In some embodiments, topics in a knowledge structure may be
represented by a concept identifier and synonyms may be represented
by a string identifier. In addition, some elements may be
represented by an alphanumeric identifier. For example, a knowledge
structure may include a map of various linked tags which point to
different elements such as, for example, web pages, pictures,
inventory, audio files, video content, etc. Additionally,
alphanumeric identifiers may correspond to temporal relationships
such as dates of conceptually linked events.
[0065] Furthermore, knowledge structures which comprise maps of
different identifiers as described above may be used by external
sources to browse concepts and retrieve information. For example, a
search engine may utilize one or more created knowledge structures
to funnel search strings into in order to acquire a particular
topic. Accordingly, search engines may improve navigation between
topics by applying their search queries to the knowledge
structures.
[0066] In some embodiments, the knowledge structure creation and
management system may enable consumers to use a computer system
over a network to contribute toward a statistically-significant or
mathematical "formalization" of synonyms or timelines for a
topic/concept. Furthermore, an exemplary knowledge structure
creation and management system may enable consumers to use a
computer system over a network to contribute toward a
statistically-significant or mathematical "formalization" of the
preferred term or phrase that should be used for a topic/concept
and for that preferred term to be able to be different by
language.
[0067] Additionally, the knowledge structure creation and
management system may enable consumers to use a computer system
over a network to contribute toward a statistically-significant or
mathematical "formalization" of relationships between
topics/concepts. Moreover, the knowledge structure creation and
management system may enable consumers to use a computer system
over a network to contribute toward the statically-significant or
mathematical "breaking" of the aforementioned terms and
relationships. Thus, some embodiments of the knowledge structure
creation and management system may enable browsing of website
content, not just searching on website content.
[0068] FIG. 8 shows an exemplary method for creating and managing
knowledge structures according to the above described general user
model. In the present example, the method may be applied to concept
searching, such as for example searching performed by a search
engine. The method may facilitate adding new concepts as well as
confirming and breaking links between concepts based on
mathematical formalization based on the preferences of the general
user populace.
[0069] The method begins at 802, where a first search query may be
received. As discussed above the search query typically may be a
text string, however it will be appreciated that the query may take
other forms.
[0070] Next at 804, the search query may be filtered through the
knowledge structures in an attempt to retrieve a related concept.
If a concept is retrieved the method moves to 806. Otherwise, a
concept is not retrieved and the method moves to 824.
[0071] Upon retrieving a concept that relates to the first search
query, a user may be presented with several knowledge structure
maintenance options. First at 806, the user may be prompted to
enter a second search query based on the retrieved concepts from
the first search query. At 808, the second search query may be
filtered through the knowledge structures in an attempt to retrieve
a related concept. If a concept is retrieved, the method moves to
810. Otherwise, the method ends.
[0072] At 810, the user may propose a link between the first search
query, the second search query, and/or the concept retrieved from
the second search query. The link may be of any suitable type such
as synonym or temporal, for example. Since this method applies to
the general user model, the proposed link may only be formalized
upon exceeding a mathematical threshold of user proposals.
[0073] Thus at 812, the number of proposals affirming the link may
be compared to a threshold. If the number of proposals exceeds the
threshold the method moves to 814. Otherwise, the method ends.
[0074] At 814, the link between the first and second search queries
and/or the retrieved concept may be formalized and the knowledge
structure may be updated.
[0075] It will be appreciated that a user may be provided the
opportunity to propose to break the link between the second search
query and the retrieved concept and may provide additional search
queries to retrieve additional matching concepts.
[0076] Returning to 804, if a concept is retrieved based on the
first search query, the method may move to 816.
[0077] At 816, the user may be prompted to break the link between
the search query and the retrieved concept. A user may choose to
break a link between two concepts if the user disagrees with the
link, for example. If the user chooses to break the link between
the first search query and the concept the method moves to 818.
Otherwise, the method ends.
[0078] At 818, the user may propose to break the link between the
first search query and the retrieved concept. Since this method
applies to the general user model, the proposed break may only be
formalized upon exceeding a mathematical threshold of user
proposals.
[0079] Thus at 820, the number of proposals to break the link may
be compared to a threshold. If the number of proposals exceeds the
threshold the method moves to 822. Otherwise, the method ends.
[0080] At 822, the link between the first search query and the
retrieved concept may be formalized and the knowledge structure may
be updated.
[0081] Returning to 804, if a concept is not retrieved based on the
first search query a user may be presented with several knowledge
structure maintenance options. First at 824, the user may enter a
second search query in an attempt to retrieve a concept via the
knowledge structure. At 826, it may be determined if a concept is
retrieved that relates to the second search query. If a concept is
retrieved, the method moves to 828. Otherwise, the method ends.
[0082] At 828, the user may propose a link between the first search
query and the second search query. In one example, the link may be
a synonym link. As another example, the link may be a temporal
link. Further, the number of proposals for the link may be compared
to a threshold at 830. If the number of proposals exceeds the
threshold the method moves to 832. Otherwise, the method ends.
[0083] At 832, the link may be formalized between the first and
second search queries and the knowledge structure may be
updated.
[0084] Returning to 804, if a concept is not retrieved based on the
first search query a user may be prompted to add a new concept
based on the search query at 834. If is determined that the user
decides to add a new concept, the method moves to 836. Otherwise,
the method ends.
[0085] At 836 the user may propose that a new concept be added to
the knowledge structure based on the first search query. The number
of proposals may be compared to a threshold at 838. If the number
of proposals exceeds the threshold the method moves to 840.
Otherwise, the method ends.
[0086] At 840, the addition of the new concept may be formalized
and the knowledge structure may be updated.
[0087] The above method may facilitate improved searching based on
concept mapping that may be continuously updated based on the
preference of the users. The general user model may enable the
knowledge structures to be refined based on the overall preference
of the general user population. In this way, concept searching and
linking concept may be improved, namely, by tailoring the
conceptual mapping as desired by the general population.
[0088] It should be understood that the knowledge structures are
relationships between concepts. The relationships may be temporal
in nature, may be semantically related (such as synonymous), and/or
a combination thereof.
[0089] FIG. 9 shows an exemplary method for creating and managing
knowledge structures according to the above described trusted user
model. In the present example, the method may be applied to concept
searching, such as for example searching performed by a search
engine. The method may be performed similarly to the method
described in FIG. 8, however adding new concepts as well as
confirming and breaking links between concepts may not be based on
mathematical formalization, but rather may be based on the
individual preference of each trusted user.
[0090] The method begins at 902, where a first search query may be
received. As discussed above the search query typically may be a
text string, however it will be appreciated that the query may take
other forms.
[0091] Next at 904, the search query may be filtered through the
knowledge structures in an attempt to retrieve a related concept.
If a concept is retrieved the method moves to 906. Otherwise, a
concept is not retrieved and the method moves to 824.
[0092] Upon retrieving a concept that relates to the first search
query, a user may be presented with several knowledge structure
maintenance options. First at 906, the user may be prompted to
enter a second search query based on the retrieved concepts from
the first search query. At 908, the second search query may be
filtered through the knowledge structures in an attempt to retrieve
a related concept. If a concept is retrieved, the method moves to
910. Otherwise, the method ends.
[0093] At 910, the trusted user may be provided with the option to
add one or more links (e.g. different relationship types) between
the first search query, the second search query, and the retrieved
concepts. If the user chooses to add a new link the method moves to
912. Otherwise, the method ends.
[0094] At 912, the link(s) between may be added between the first
search query, the second search query, and/or the retrieved
concepts. The link(s) may be formalized immediately since the
trusted user model does not require additional reinforcement from
the user populace.
[0095] Returning to 904, upon retrieval of a concept based on the
first search query, the trusted user may be provided the option to
break the line between the concept and the first search query at
904. If the trusted user chooses to break the link the method moves
to 916. Otherwise, the method ends.
[0096] At 916, the link between the first search query and the
retrieved concept may be broken (or removed). The change in the
knowledge structure may be made immediately since the trusted user
model does not require additional reinforcement from the user
populace.
[0097] Returning to 904, if a concept is not retrieved based on the
first search query a user may be presented with several knowledge
structure maintenance options. First at 918, the user may be
prompted to enter a second search query in an attempt to retrieve a
concept via the knowledge structure. At 920, it may be determined
if a concept is retrieved that relates to the second search query.
If a concept is retrieved, the method moves to 922. Otherwise, the
method ends.
[0098] At 922, it may be determined if the user chooses to add a
link (e.g. a temporal link and/or a synonym type link) between the
first search query and the second search query. If the user chooses
to add a link the method moves to 924. Otherwise, the method
ends.
[0099] At 924, the link between the first search query and the
second search query may be formalized immediately since the trusted
user model does not require additional reinforcement from the user
populace.
[0100] Returning to 904, if a concept is not retrieved based on the
first search query a trusted user may be provided with the option
to add the first search query as a new concept in the knowledge
structure. If the trusted user chooses to add a new concept the
method moves to 928. Otherwise the method ends.
[0101] At 928, the first search query may be added as a new concept
to the knowledge structures and may be formalized immediately since
the trusted user model does not require additional reinforcement
from the user populace.
[0102] The above method may facilitate improved searching based on
concept mapping that may be continuously updated based on the
preference of the users. The trusted user model may enable the
knowledge structures to be refined based on the preferences of each
of the trusted users. As discussed above, the trusted user model
may make use of the individual expertise of each trusted user in
various subjects to improve the concept mapping of the knowledge
structures. In this way, concept searching and linking concept may
be improved, namely, by tailoring the conceptual mapping as desired
by each of the trusted users.
[0103] Furthermore, it will be appreciated that in some embodiments
the methods described above and shown in FIG. 8 and FIG. 9 may be
combined for a hybrid model. In such a model, trusted users may
modify the knowledge structures immediately whereas general users
may make proposed changes that require a threshold amount to
formalize the changes to the knowledge structures.
[0104] FIG. 10 shows an exemplary embodiment of a concept search
graphical user interface 1000 that may be generated by a knowledge
structure creation and management system. It should be appreciated
that the illustrated embodiment is purely exemplary and the
knowledge structure creation and management system may include
other presentation formats without departing from the scope of the
disclosure. For example, the search interface may include separate
pages for each concept, a plurality of drop down menus that include
various search terms, and other suitable components pertaining to
the search process.
[0105] As shown at 1002, a first example section shows a search
queried for the concept of "Madonna". The knowledge structure
creation and management system may include a drop down menu
offering a variety of concepts which relate to the search term. A
first search path (or concept map) 1004 shows that the concept of
"Art and Pop Culture" includes a sub-type of "Pop Music" which
includes the relationship type "Performers of" which is linked to
the concept "Madonna". A second search path 1006 includes the
concept "Religion and Faith" which includes a sub a sub-type of
"Christianity" which includes the relationship type "Symbols of"
which is linked to the concept "Madonna".
[0106] Furthermore, the concept "Madonna" may be linked to other
concepts through other relationship types. For example, as shown at
1008, the concept "Madonna" includes the relationship type
"Performer of" which links to the concept "Evita" which is a
sub-type of "Movies" and which is additionally linked by the
sub-type "1996".
[0107] Additionally, as shown in a second example section 1010, the
concept "Evita" is a sub-type of "Musicals which is a sub-type of
"Art and Pop Culture". The concept "Evita" further may be linked to
other concepts via different relationship types, namely, "Composers
of", "Performers of", "Currently Playing", and "Related Topics" for
example.
[0108] The illustrated embodiment further includes a third section
1012 that may include "Boards" and "New Concepts" portions (e.g.
thumbnails) which include various concepts and sub-types of
concepts. In some embodiments, the "Boards" portion may provide a
list of the most searched concepts. In some embodiments, the "New
Concepts" portion may provide recently added concepts and/or links
between concepts.
[0109] As described above, some knowledge structures may be
organized such that concepts may be linked temporally. As one
example, concepts may be organized in chronological representations
(e.g. timelines). The chronological representations may detail all
related events including large events and sub-events that pertain
to the focused (or searched) event. By organizing the different
events into chronological representations the presentation of
temporally linked concepts may be configured in an intuitive
format.
[0110] The temporal knowledge structures may be modified as
discussed above using general user, trusted user and hybrid models.
In particular, a user, when combined with other users, may be able
to contribute to the linkages between events. By the same token, a
user might be able to challenge a relationship, and with other
user's challenges may break the temporal relationship (or link)
between events.
[0111] Furthermore, it will be appreciated that the knowledge
structures may be searched according to both concept and time. In
particular, temporal searching may be conducted by changing the
granularity of a timeline to search for events over different or
varying periods of time. Further, retrieved events may also have
relationships with different concepts which may be presented to the
user.
[0112] FIG. 11 shows an exemplary embodiment of a temporal search
graphical user interface 1100 that may be generated by a knowledge
structure creation and management system. In the illustrated
embodiment, a user may enter search queries into a text field and
the query may be applied to the temporal knowledge structures to
return a concept in the search results field 1102. The search
results 1102 may return a main concept or event and further may
retrieve sub-concepts 1104 linked to the main concept. In some
embodiments, the sub concepts may be retrieved based on the
strength of the link between the concepts established by the user
of the knowledge structure creation and management system.
[0113] The temporal search graphical user interface 1100 may be
configured to present a temporal representation (e.g. timeline)
1106 that corresponds with the search results 1102. The temporal
representation 1106 may include a plurality of sub-events (or
sub-concepts) that may be mapped to the main concept. In the
illustrated embodiment, sub concepts may be depicted as circles on
the timeline. Each of the sub-events may include an event
description 1108 and a time stamp 1110. Further, the temporal
representation 1106 may include concepts that do not necessarily
relate to the main concept or may have a statistically weak
relationship with the main concept, yet take place in the span of
the temporal representation (e.g. timeline). In the illustrated
embodiment, the conceptually unrelated event 1112 is depicted as a
square on the timeline.
[0114] The temporal search graphical user interface 1100 may
include a zoom adjustment 1114 depicted as a scrolling bar. The
zoom adjustment may be used to control the length the timeline or
the granularity of events on the timeline. In the illustrated
example, the length of the timeline is set for three hours spanning
from 8:00 AM to 11:00 AM on Sep. 11, 2001. Further, the granularity
may be set according to the amount of time with events being
displayed based on their relationship with the main concept. That
is to say, events with strong links to the main concept based on
user preference may be displayed over other concepts where
applicable.
[0115] The zoom adjustment 1114 may be modified to increase or
decrease the length of the timeline and the events displayed on the
timeline may be adjusted to reflect the strength of the link
between concepts. For example, the timeline may be zoomed in to a
length of one hour which may eliminate certain events from being
displayed that do not fall within the particular time span.
Further, additional events may be displayed to account for the
adjustment in granularity.
[0116] The temporal search graphic user interface 1100 may include
a detail display tool 1118 that may be configured to display
additional event information 1120 upon selecting an event on the
timeline. The display tool may generate a pop-up display that
presents event information as well as system management
information. For example, event information 1120 may include
different timeline elements, such as for example, the title of the
event, a description of the event, the time and date of the event,
conceptual links to other events, and pending system actions on the
event. Pending actions on the event may correspond to the
management methods described in FIG. 8. In one particular example,
a timeline element for a particular event may present the number of
challenges to the time of the event, the number of challenges to
the link between the event and the main concept, and/or the number
of affirmations of the link between the event and the main
concept.
[0117] The temporal nature of the knowledge structures may be
presented in an intuitive manner that facilitates quick and
flexible searching due to the vast number of concepts and event
that are linked through various relationships.
[0118] It will be appreciated that the search aspect of the
graphical user interface of FIG. 11 is exemplary. Further, the
temporal knowledge structures may be applied to other suitable
applications without leaving the scope of the present disclosure.
For example, the temporal knowledge structures may be applied to
operation logistic, medical history, etc.
[0119] Furthermore, it will be appreciated that the temporal
knowledge structures may be created and managed according to
methods of FIGS. 8 and 9. In particular, the general user and
trusted user models may be implemented to modify concepts and
relationships between concepts in the knowledge structures as
discussed above.
[0120] It will be appreciated that the above described knowledge
structure creation and management system and various user models
may be integrated into other systems. In one example, the knowledge
structure creation and management system (using temporal and/or
semantic knowledge structures) may be adapted for use by websites
and may be integrated into search systems to improve searching. In
particular, the knowledge structure creation and management system
may enable a website to embed topic-based navigation and search
applications into the website. In an exemplary embodiment of the
knowledge structure creation and management system, users may
navigate a website using temporal navigation and/or topic-based
navigation that is dynamically built from relevant strings from
various websites, synonymous strings mapped to a specific topic,
temporal strings, and/or inter-topic navigation.
[0121] Furthermore, knowledge structure management tools may be
embedded into website navigation. For example, in some embodiments
a user may be prompted to "vote" to affirm or decline concepts
returned during a search or to confirm or break links between
concepts returned during a search. In one particular example, a
user may "vote" to confirm a temporal link between two
concepts.
[0122] In some embodiments, different knowledge structures may
comprise mappings which include elements from various different
languages. Furthermore, a single concept may be linked to
descriptive character strings from any desired language within a
knowledge structure. In such a configuration, conceptual
organization may facilitate improved information retrieval and
topic browsing, since the system may transcend barriers created by
dependence on a single language.
[0123] Although exemplary embodiments of the knowledge structure
creation and management system have primarily been associated with
searching computer systems over a network; it should be appreciated
that some embodiments of the knowledge structure creation and
management system may be applied to a variety of different
applications and environments. For example, in a marketing
environment, the knowledge structure creation and management system
may facilitate advertising to be sold based on one or more concepts
or conceptual links which may be applied in cross product
advertising. Furthermore, the knowledge structure creation and
management system may facilitate advertising to be sold or rendered
based on groups of related concepts.
[0124] Another example application of a knowledge structure may be
applied to law. In an exemplary application, the knowledge
structure creation and management system may include trusted users
in the form of legal scholars to collaboratively establish the
ontology of federal and state laws. Furthermore, the knowledge
structure creation and management system may enable general users
(i.e. the public) to navigate to lawyers based on practice
specialty. For example, the concept of lawyers may be searched
according to a sub-type of relationship, for example real estate
law, which may link to the concept of landlord-tenant privacy
disputes.
[0125] Another example application of a knowledge structure may be
applied to automobiles. In an exemplary application, the knowledge
structure creation and management system may include trusted users
in the form of consumers who are experts in the automobile and auto
parts industry to collaboratively establish a repair or replacement
parts catalog for classic automobiles. Once the catalog is
established, the catalog may be used by auto parts manufacturers to
better classify and offer these parts to consumers.
[0126] Another example application of a knowledge structure
creation and management system may be applied to entertainment. In
an exemplary application, the knowledge structure may enable
consumers to create and confirm an ontology of all artists and
their works, allowing a consistent and complete catalog of topics
to drive product sales and information retrieval.
[0127] Another example application of a knowledge structure
creation and management system may be applied to record keeping. In
particular, events may be confirmed in a temporal knowledge
structure allowing for an overview of a lifetime of someone or
something. More particularly, the temporal knowledge structures may
be applied to individuals, businesses, or other entities.
[0128] Furthermore, although many examples include searching
knowledge structures for information, it should be appreciated that
an exemplary knowledge structure system may include functions to
create, manage, and remove knowledge structures. Furthermore, it
should be appreciated that searching may be a tool utilized in the
management of an exemplary knowledge structure system.
[0129] The above described applications and environments in which
embodiments of the knowledge structure creation and management
system may be implemented should be considered purely exemplary. It
should be appreciated that the knowledge structure creation and
management system may be used to search various systems and
concept, as well as take various different formats and
configurations without departing from the scope of this disclosure.
For example, the knowledge structure creation and management system
may be used to search concepts across different subject such as,
standing inventories, historical events, sporting statistics,
environmental conditions, video content, mathematic equations,
etc.
[0130] While the present invention has been particularly shown and
described with reference to the foregoing preferred embodiments,
those skilled in the art will understand that many variations may
be made therein without departing from the spirit and scope of the
invention. The description of the invention should be understood to
include all novel and non-obvious combinations of elements
described herein, and claims to any novel and non-obvious
combination of these elements.
* * * * *