U.S. patent application number 12/958785 was filed with the patent office on 2011-06-30 for method and system for social network analysis.
This patent application is currently assigned to eBay Inc.. Invention is credited to Zeqian Shen, Neelakantan Sundaresan.
Application Number | 20110161191 12/958785 |
Document ID | / |
Family ID | 40433064 |
Filed Date | 2011-06-30 |
United States Patent
Application |
20110161191 |
Kind Code |
A1 |
Shen; Zeqian ; et
al. |
June 30, 2011 |
METHOD AND SYSTEM FOR SOCIAL NETWORK ANALYSIS
Abstract
Methods and system for social commerce network analysis are
described. In one embodiment, user interaction data associated with
users for a time period in a social commerce network is accessed
and a network analysis is performed. Users within the social
commerce network may be selected where each of the users is
associated with reputation information. A motif for the users for
the time period based on the network analysis may be generated. A
node of the motif is associated with a particular user, and the
motif defines an expected relationship between the users in the
social commerce network. The node is distinguished based on the
reputation information of the user. The motif may be used with a
plurality of distinguished nodes for analysis of the social
commerce network.
Inventors: |
Shen; Zeqian; (San Jose,
CA) ; Sundaresan; Neelakantan; (Mountain View,
CA) |
Assignee: |
eBay Inc.
San Jose
CA
|
Family ID: |
40433064 |
Appl. No.: |
12/958785 |
Filed: |
December 2, 2010 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11967221 |
Dec 30, 2007 |
|
|
|
12958785 |
|
|
|
|
60971904 |
Sep 12, 2007 |
|
|
|
60984677 |
Nov 1, 2007 |
|
|
|
Current U.S.
Class: |
705/26.1 ;
709/224 |
Current CPC
Class: |
G06Q 30/0601 20130101;
G06Q 30/02 20130101; G06Q 50/01 20130101 |
Class at
Publication: |
705/26.1 ;
709/224 |
International
Class: |
G06Q 30/00 20060101
G06Q030/00; G06F 15/173 20060101 G06F015/173 |
Claims
1. A method comprising: accessing user interaction data associated
with a plurality of users for a time period in a social commerce
network; performing network analysis, using one or more processors,
on the user interaction data; selecting a plurality of example
users within the social commerce network, each of the plurality of
example users being associated with reputation information;
generating a motif for the plurality of example users for the time
period in accordance with the performing of the network analysis, a
node of the motif being associated with a particular example user
of the plurality of example users, the motif defining an expected
relationship between the plurality of example users in the social
commerce network; distinguishing the node of the plurality of
example users in accordance with the reputation information of a
respective example user; and utilizing the motif with a plurality
of distinguished nodes for analysis of the social commerce
network.
2. The method of claim 1, wherein the utilizing comprises:
providing the motif with the plurality of distinguished nodes for
presentation.
3. The method of claim 1, wherein the utilizing comprises:
analyzing the motif and a plurality of additional motifs; and
making a decision regarding the social commerce network in
accordance with the analyzing of the motif.
4. The method of claim 1, further comprising: accessing interaction
frequency data associated with the plurality of users; and applying
a texture to at least one connecting line of the motif in
accordance with the interaction frequency data.
5. The method of claim 1, further comprising: accessing
transactional financial data associated with the plurality of
users; and applying a texture to at least one connecting line of
the motif in accordance with the transactional financial data.
6. The method of claim 1, wherein the distinguishing comprises:
colorizing the node of the plurality of example users in accordance
with the reputation information.
7. The method of claim 1, wherein the social commerce network
comprises a plurality of buyers and a plurality of sellers.
8. The method of claim 1, further comprising identifying a category
and wherein the accessing of the user interaction data is based on
the identified category.
9. The method of claim 1, wherein the motif is a triad.
10. A non-transitory machine-readable medium comprising
instructions, which when implemented by one or more processors
perform operations comprising: accessing user interaction data
associated with a plurality of users for a time period in a social
commerce network; performing network analysis, using the one or
more processors, on the user interaction data; selecting a
plurality of example users within the social commerce network, each
of the plurality of example users being associated with reputation
information; generating a motif for the plurality of example users
for the time period in accordance with the performing of the
network analysis, a node of the motif being associated with a
particular example user of the plurality of example users, the
motif defining an expected relationship between the plurality of
example users in the social commerce network; distinguishing the
node of the plurality of example users in accordance with the
reputation information of a respective example user; and utilizing
the motif with a plurality of distinguished nodes for analysis of
the social commerce network.
11. The non-transitory machine-readable medium of claim 10 further
comprising additional instructions, which when implemented by the
one or more processors perform operations comprising: accessing
interaction frequency data associated with the plurality of users;
and applying a texture to at least one connecting line of the motif
in accordance with the interaction frequency data.
12. The non-transitory machine-readable medium of claim 10, wherein
the utilizing comprises: providing the motif with the plurality of
distinguished nodes for presentation.
13. The non-transitory machine-readable medium of claim 10, wherein
the utilizing comprises: analyzing the motif and a plurality of
additional motifs; and making a decision regarding the social
commerce network in accordance with the analyzing of the motif.
14. The non-transitory machine-readable medium of claim 10 further
comprising additional instructions, which when implemented by the
one or more processors perform operations comprising: accessing
transactional financial data associated with the plurality of
users; and applying a texture to at least one connecting line of
the motif in accordance with the transactional financial data.
15. The non-transitory machine-readable medium of claim 10, wherein
the distinguishing comprises: colorizing the node of the plurality
of example users in accordance with the reputation information.
16. The non-transitory machine-readable medium of claim 10, wherein
the social commerce network comprises a plurality of buyers and a
plurality of sellers.
17. The non-transitory machine-readable medium of claim 10, further
comprising identifying a category and wherein the accessing of the
user interaction data is based on the identified category.
18. A method comprising: accessing reputation information
associated with a plurality of initiating users and a plurality of
responding users in a social network for a time period; accessing
interaction frequency data associated with the plurality of
initiating users and the plurality of responding users for the time
period; plotting an aggregated correlation between the plurality of
initiating users and the plurality of responding users in
accordance with the reputation information; differentiating, using
one or more processors, the plotting of the aggregated correlation
in accordance with the interaction frequency data; and utilizing
the differentiated plotting of the aggregated correlation.
19. The method of claim 18, wherein the utilizing comprises:
providing the differentiated plotting of the aggregated correlation
for presentation.
20. The method of claim 18, wherein the utilizing comprises:
accessing the reputation information associated with a plurality of
assorted initiating users and a plurality of assorted responding
users in the social network for an additional time period;
accessing interaction frequency data associated with the plurality
of assorted initiating users and the plurality of assorted
responding users for the additional time period; plotting the
aggregated correlation between the plurality of assorted initiating
users and the plurality of assorted initiating users in accordance
with the reputation information; and using the differentiated
plotting of the aggregated correlation for the time period and the
additional time period for analysis of the social network.
Description
[0001] This application is a continuation application and claims
priority benefit of U.S. patent application Ser. No. 11/967,221
filed Dec. 30, 2007 and entitled "Method and System for Social
Network Analysis" which, in turn, claims the benefit of both U.S.
Provisional Patent Application Ser. No. 60/971,904, entitled
"Social Network Analysis," filed Sep. 12, 2007 and U.S. Provisional
Patent Application Ser. No. 60/984,677, entitled "Analysis of a
Social Commerce Network," filed Nov. 1, 2007; the entire contents
of which are herein incorporated by reference.
TECHNICAL FIELD
[0002] The present application relates generally to the technical
field of data analysis and management and, in one specific example,
to social commerce network analysis.
BACKGROUND
[0003] The Internet and World Wide Web include content spaces,
commerce spaces, and spaces for social interactions. Within the
spaces for social interactions, users may interact with one another
in both commercial settings and non-commercial settings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] Some embodiments are illustrated by way of example and not
limitation in the figures of the accompanying drawings in
which:
[0005] FIG. 1 is a block diagram of an environment in which various
embodiments described herein may be practiced.
[0006] FIG. 2 is a flowchart depicting a method for plotting a
degree distribution, according to some embodiments.
[0007] FIG. 3 is an example plot of a degree distribution,
according to some embodiments.
[0008] FIG. 4 is a second example of a plot of a feedback
distribution, according to some embodiments.
[0009] FIG. 5 includes additional examples of degree distribution
and feedback distributions within categories in the social commerce
network.
[0010] FIG. 6 is a flowchart depicting a method for determining a
social strength of a network, according to some embodiments.
[0011] FIGS. 7-11 are example graphs that may be generated using
the method of FIG. 6.
[0012] FIG. 12 is an example table that may be generated using the
method of FIG. 6, according to an example embodiment.
[0013] FIG. 13 is an example bar graph comparing how the social
strength of the social commerce network changes on a per-category
basis.
[0014] FIG. 14 is an example graph of relative social strength
across a number of categories.
[0015] FIG. 15 depicts examples of motifs, according to some
embodiments.
[0016] FIG. 16 is an example graph based on a motif analysis
according to some embodiments.
[0017] FIG. 17 illustrates a method for conducting motif analysis
according to an example embodiment.
[0018] FIG. 18 is an example motif display according to an example
embodiment.
[0019] FIG. 19 is a flowchart of a method for differentiated
plotting analysis according to an example embodiment.
[0020] FIG. 20 depicts example Auroral diagrams generated according
to some embodiments.
[0021] FIG. 21 is a block diagram of a machine in the example form
of a computer system within which instructions, for causing the
machine to perform any one or more of the methodologies discussed
herein, may be executed.
DETAILED DESCRIPTION
[0022] Example methods and systems to analyze social commerce
networks are described. In the following description, for purposes
of explanation, numerous specific details are set forth in order to
provide a thorough understanding of example embodiments. It will be
evident, however, to one skilled in the art that the inventive
subject matter may be practiced without these specific details.
[0023] Social scientists have extensively studied offline social
commerce networks and have developed measures for connectivity,
trust, and evolution. As the web evolves from being a content and
commerce space to a space of social interactions, study of the
nature of these social interactions has become a key area for
Internet researchers. Gaining a deeper understanding of online
social networks in terms of what structure they have, what kind of
interactions they afford, what trust and reputation measures exist,
and how they evolve may allow one to build effective user
applications and business models around these networks. When social
networks have commerce in them even the simplest events like search
or mouse clicks can have tangible revenue measures associated
therewith.
[0024] Some embodiments described herein may be used to study
social networks driven by commerce, such as eBay.RTM.. As a large
online auction site, eBay.RTM. has been studied in various areas,
including auction models, bidding and selling strategies and
reputations models. However, there is no previous study on its
social commerce network. The social commerce network is a social
network with entities connecting by, for example, trading relations
between them.
[0025] In addition, it may be desirable to model and analyze a
social commerce network as a complex network, because of its
large-scale and complex topological features. Complex networks
analysis methods focus on large-scale statistical properties of
networks. Bow-tie structural analysis provides a way to model the
macroscopic structure of networks.
[0026] For characterization of microscopic structure within the
social commerce networks, significant recurring patterns, referred
to as network motifs, may be determined. Motif profiling has been
conducted on various networks including genetic networks, neuronal
wiring, World Wide Web, social commerce networks, and
word-adjacency networks. The networks are clustered into several
super-families based their motif significance profiles.
[0027] In the examples used to illustrate some embodiments, a
dataset of about 78 million transactions of the eBay.RTM.
Marketplace involving about 14.5 million users was used. The
trading network derived from the dataset is a directed graph. Each
node denotes a user. An edge from node A to node B exists, if A
sold something to B at least once. The number of transactions
between the same seller and buyer is stored as an edge weight. A
vertex in the network can be a seller in some transactions and a
buyer in others. There might be bidirectional edges between two
vertices.
[0028] The dataset also captured the listing titles, listing
categories, and user feedback scores of each transaction. Using the
category information, sub-networks for individual categories may be
derived. The data used spans over 32 top-level categories. The
categories are categories of products listed for sale such as
Autos, Baby, Collectibles, Jewelry and Watches, Computers and
Networking, Crafts, Antiques, etc.
Architecture
[0029] FIG. 1 illustrates an example system 100 in which a
community of users may use a number of client machines 102 to be
involved in a social network. The client machine 102 may be a
computing system, a mobile phone, a personal digital assistant
(PDA), a gaming unit, a portable computing unit, and the like. The
social network may be a social commerce network over which the
users operating the machines may be involved in commercial exchange
(e.g., buying or selling). However, other types of social networks
(e.g., informational social networks) may also be used.
[0030] In an example embodiment, the social commerce network may be
a social structure made of nodes (e.g., users, such as individuals
or organizations) that are tied by one or more specific types of
interdependency including, by way of example, values, visions,
ideas, commerce, friends, kinship, dislike, conflict, web links,
sexual relations, disease transmission, or airline routes. For
example, a social commerce network may be a network that includes a
commercial interdependency.
[0031] The client machines 102 may participate in the social
commerce network by communicating over a provider network 104 with
a network analyzer 106. The provider network 104 may be a Global
System for Mobile Communications (GSM) network, an Internet
Protocol (IP) network, a Wireless Application Protocol (WAP)
network, a WiFi network, or an IEEE 802.11 standards network as
well as various combinations thereof. Other conventional and/or
later developed wired and wireless networks may also be used.
[0032] The network analyzer 106 allows the social commerce network
to be provided to the users of the client machines 102. The network
analyzer 106 may be used to analyze the social commerce network by
using a degree distribution subsystem 108, a social strength
subsystem 110, a motif subsystem 112, and a preferential
connections subsystem 114. The degree distribution subsystem 108 is
to analyze a social commerce network using degree distribution, a
measure for complex networks. The social strength subsystem 110 is
to determine a social strength of the social commerce network by
calculating the various levels of contribution of the members of
the social commerce network. The motif subsystem 112 is used to
analyze recurring, small, connected sub-networks within the social
commerce network. The preferential connections subsystem 114 is
used to quantify users' tendencies to interact (e.g., participation
in a transaction) with other users based on their feedback
scores.
[0033] A social commerce network may be directed, meaning that
edges point in one direction from one node to another node (e.g.,
from a seller to a buyer). In a directed network, the nodes may
each have two different degrees, the in-degree, which is the number
of incoming edges (e.g., where the user is a buyer); and the
out-degree, which is the number of outgoing edges (e.g., where the
user is a seller).
[0034] FIG. 2 is a flowchart of a degree distribution method 200 to
determine the degree distribution of the social commerce network.
The degree distribution method 200 may be performed by the social
strength subsystem 110 (FIG. 1). The degree of a node (the node
representing, e.g., a user) in the social commerce network is the
number of connections it has to other nodes. The degree
distribution is a probability distribution of these degrees over
all of the nodes within a network. Many complex networks exhibit a
scale-free nature characterized by the power law distribution of
node degrees, P(k).about.k.sup.-r, where P(k) is the fraction of
vertices (e.g., nodes) in the network that have degree k.
[0035] In an operation 202, a particular node is selected. The
particular node may be identified randomly or according to a
sorting method. In some instances, the nodes may be identified by a
number assigned to the node. The node may represent a member of the
community, such as a member, a user, a buyer, a seller, etc.
[0036] In an operation 204, the in-degree of the node is determined
and, in an operation 206, the out-degree of the node is determined.
The in-degree or the out-degree may represent the number of
transactions engaged in by the user represented by the node.
Specifically, the in-degree may be a count of transactions where
the user is a buyer and the out-degree may be a count of
transactions where the user is a seller. Each of operations 202,
204, and 206 is repeated for each node in the network. In an
operation 208, the probability distribution is calculated over the
nodes. In an operation 210, the probability distribution is
plotted, as shown in FIG. 3.
[0037] The social commerce network is scale-free, as depicted by
the plot 300 in FIG. 3. In the case of out-degree, the calculated
value of the exponent of the power law is around 1.7 based on the
exemplary data plotted. The exponent for in-degree distribution is
around 2.98. The results are similar to the distribution observed
for other complex networks like the Web. The number of users with
small out-degree is far fewer than the number of those with the
same value of in-degree. As the value increases, the number of
users with certain out-degree exceeds those with the same
in-degree. The critical point for this network is about 130. Put
another way, sellers in this network tend to be more experienced
than buyers.
[0038] The degree distribution method 200 may be used to analyze
other aspects of the social commerce network, as depicted by FIG.
4. FIG. 4 depicts a feedback score distribution 400. In order to
plot the log-scaled figure, users with negative feedbacks are
dropped. The X axis shows the scores and the Y axis shows the count
of users corresponding to these feedback scores. The feedback score
distribution 400 also exhibits the power law phenomenon at the
exponent of around 2.05.
[0039] FIG. 5 includes additional examples of degree distribution
and feedback distributions of sub-networks of twelve representative
categories in the social commerce network. Most of the
distributions also exhibit the power-law phenomenon. Generally,
distribution of feedback scores are more skewed at the lower end
than degree distributions.
[0040] FIG. 6 is a flowchart of a method 600 for determining
strength of the social commerce network. The strength of the social
commerce network may be understood as the number or percentage of
members who are actively involved in the social commerce network
based on user interaction data. The user interaction data may
include data collected over a time period for a number of users in
a social commerce network and/or include additional user
interaction data associated with the social commerce network during
a different time period. The user interaction data may be based on
communications between users. For example, the user interaction
data may include, by way of example, transactional relationship
data that relates to a transaction (e.g., a sale or item exchange)
that has occurred between users, communication interaction data
that relates to a communication (e.g., an e-mail, an instant
message, or a voice over IP call) that has occurred between users,
and the like.
[0041] A complex network can be separated into six parts: strongly
connected component (SCC), in component (IN), out component (OUT),
Tendrils, Tubes, and Disconnected. The SCC is the maximal strongly
connected component, in which for every pair of nodes u and v,
there is a path from u to v and a path from v to u. Nodes from
which one reaches the SCC form the IN component, while the OUT
component consists of nodes reached from the SCC. There are paths
from IN to OUT without going through the SCC. The nodes along these
paths form another group called Tubes. The Tendrils component
gathers nodes that either can reach the OUT component or are
reachable from the IN component, but do not belong to any the other
defined regions. The Disconnected component consists of a few nodes
that are disconnected to these components.
[0042] In the example social commerce network, the SCC contains
users that frequently dealt with each other directly or indirectly
and may be referred to as "active traders." If a user sells
something to a user in the SCC, it may be bought by anyone else in
the SCC. The IN component contains users that usually only sell
things to the SCC. They could be power members(sellers) of the
community. The OUT component contains users that usually only buy
things from the SCC. They could be casual or power members(buyers)
of the network. Users in Tendrils and Tubes are users only dealt
with IN and OUT, but not the SCC. There are also "Disconnected"
users, who did not deal with all the users mentioned above.
[0043] Referring back to FIG. 6 that illustrates a method 600
according to an example embodiment, the method 600 may be performed
by the social strength subsystem 110 (FIG. 1).
[0044] User interaction data of a time period for users in a social
network (e.g., a social commerce network) is accessed at block 602.
The user interaction data may be associated with a single
transaction category or multiple transaction categories. In some
instances, user interaction data collected over two or more time
periods may be accessed and compared using method 600.
[0045] Network analysis is performed on the user interaction data
at block 604. A graph is generated from the user interaction data
in accordance with the performing of the network analysis at block
606. The graph may be a necktie-shaped graph, a bowtie-shaped
graph, or be a different shape based on the characteristics of the
user interaction data. In addition, the graphs may have different
sized dimensions based on a particular representation.
[0046] At decision block 608, a determination may be made whether
to apply a texture to the graph. If a determination is made to
apply a texture, additional user data may be accessed at block 610
and a texture may be applied to the graph in accordance with the
additional user data at block 612. The additional user data may
include, by way of example, interaction frequency data, reputation
information, transactional financial data, or other data associated
with the users.
[0047] The texture applied to the graph may include colorization,
striping, and the like. The texture may better enable a user to
more easily understand more about the relationships of the users
reflected in the graph. For example, users in a particular
component of the graph may be making a large number of interactions
or a small number of interactions.
[0048] If a determination is made at decision block 608 not to
apply a texture or upon completion of the operations at block 612,
the method 600 may proceed to block 614.
[0049] The graph is utilized for analysis of the social commerce
network at block 614. For example, the graph may be provided for
presentation.
[0050] FIG. 7 is a diagram of an example necktie-shaped graph 700
according to an example embodiment. The necktie-shaped graph 700 is
an example representation of a graph that may be generated in
accordance with the method 600. However, other representations of
the same or different types of graphs may also be generated.
[0051] A strongly connected component 702 may be connected to an
in-component 704 and an out-component 706. While the representation
of the example necktie-shaped graph 700 reflects a strongly
connected component value of 5.83%, an in-component value of 3.03%,
and an out-component value of 65.83%, other values may be used in
other representations.
[0052] The in-component 704 and the out-component 706 may be
connected through a tube 708. The tube value in the presentation
representation is 0.64%, however other values may be used in other
representations.
[0053] One or more tendrils 710 may be connected to the
in-component 704 or the out-component 706. However, a different
number of tendrils 710 may be used in a different representation.
The tendril value in the presentation representation is 23.59%,
however other values may be used in other representations.
[0054] A disconnected component 712 may not be connected to the
strongly connected component 702, the in-component 704, the
out-component 706, the tube 708, and/or the tendrils 710. The
disconnected component value in the presentation representation is
1.09%, however other values may be used in other
representations.
[0055] In order to find out the characteristic of each component,
the feedback scores, number of purchase and sales of each component
in the social commerce network are measured. Users in the SCC are
active with an average of 14.16 purchases and 55.83 sales. Users in
SCC sell more than buy. Users in IN are mainly sellers, who made
63.84 sales while only 0.42 make purchases on average. Users in OUT
are buyers, who made 6.48 purchases and only 0.21 sales on average.
Also the users in OUT are typically casual users. The SCC and IN
have the highest average feedback scores. The users in the tubed
708 have relatively high feedback scores. These are experienced
users.
[0056] FIGS. 8-11 are diagrams of example graphs 800, 900, 1000,
and 1100 according to example embodiments. Each of the graphs 800,
900, 1000, and 1100 include a strongly connected component, an
in-component, an out-component, a tube, one or more tendrils, and a
disconnected component. The graph 800 may represent an Antiques
category, the graph 900 may represent a Collectibles category, the
graph 1000 may represent a Sports Memorabilia and Cards category,
and the graph 1100 may represent a Stamps category.
[0057] FIG. 12 illustrates a block diagram of an example table 1200
generated using the method 600 (see FIG. 6). The table 1200 is an
example representation that may reflect a distribution of various
components of a graph (e.g., the necktie-shaped graph 700 of FIG.
7). However, other representations containing different values
and/or components may also be used.
[0058] For example, the size row 1202 of the table 1200 may reflect
a size percentage of various components of a graph. The average
feedback row 1204 may contain values that reflect an average
feedback store of users associated with a particular component. The
purchases row 1206 may contain values that reflect an average and
percentage of purchases associated with a particular component. The
sales row 1208 may contain values that reflect an average and
percentage of sales associated with a particular component. The
columns 1210-1220 are associated with particular components of a
graph.
[0059] FIG. 13 is an example bar graph 1300 comparing how the
social strength of the social commerce network changes on a
per-category basis. First, the size of SCC, which indicates the
size of active community in each category, is compared. The average
percentage of SCC is 0.61%, and the standard deviation is 0.8%. The
categories with relatively larger SCC (more than one standard
deviation larger than the average value) are Toys & Hobbies,
Stamps, Sports Memorabilia, Cards & Fan Shop, Dolls &
Bears, Crafts, Collectibles, Clothing, and Shoes & Accessories.
Most of them are collectible merchandises, with which strong
communities of collectors are more often formed. The five
categories with the smallest SCC are Specialty Services, Live
Auctions, Travel, Consumer Electronics and Real Estate. These
categories either have large Tendrils or Disconnected components.
They are more like traditional retail networks, in which entities
are either buyers or sellers. There are few trading relations
between peer buyers or sellers.
[0060] In addition, a measurement indicating how asymmetric the
bow-tie structures are based on the ratio of size of OUT over size
of IN may be determined (see, for example FIGS. 7-11). The average
value of the ratio is 7.25 with a standard deviation of 6.25. The
categories with large OUT-IN ratios are Everything Else (30.0),
Health & Beauty (15.5) and Home & Garden (20.4). These
Categories have much more active buyers than active sellers. Baby,
Stamps, Specialty Services, Travel, and Live Auctions are the
categories whose IN component is larger than OUT component.
[0061] Next, the individual categories with the social commerce
network may be compared. The SCC, IN, and OUT components of the
social commerce network are aggregations of corresponding
components in the sub-networks. When two sub-networks are merged,
the new SCC is at least the union of two SCCs if those two
intersect with each other. In addition, vertices in IN and OUT can
also become part of the SCC.
[0062] Therefore, the percentage of SCC in the social commerce
network increases. The Tendrils component is much less significant
than those of individual categories. Since the IN component does
not increase significantly, most of the vertices moved from the
Tendrils into the OUT component. In other words, a large number of
casual users in Tendrils of a category are likely to be active
buyers in some other categories.
[0063] Social commerce is used to build communities around common
commercial interests, such that transactions and revenues can be
maximized. Therefore, the social strength of the networks is
measured. In the tie model of FIG. 7-11, the SCC component is the
closely connected center community, which includes the most active
users. The IN and OUT are power members. The Tendrils and Tube are
relatively casual members. Different components show different
levels of social participation and trading activities. In other
words, they have different level of contribution to the total
social strength. Therefore, the measurement for the social strength
of a social commerce network is based on its tie structure. The
measurement is a linear combination of proportions of the
components except the Disconnected component, because there is few
activities in this part. For example, the social strength of a
social commerce network may be calculated by:
SocialStrength=a.sub.1*SCC+a.sub.2*IN+a.sub.3*OUT+a.sub.4*Tube+a.sub.5*T-
endrils
where the coefficients a.sub.1, a.sub.2, a.sub.3, a.sub.4, and
a.sub.5 are the weights on the component. The coefficient may be
used to express how much a vertex in each component can contribute
to the total social strength. Therefore, it may be desirable to
select the coefficients so that:
a.sub.1>a.sub.2=a.sub.3=a.sub.4>a.sub.5.
For example, {1.0, 0.5, 0.5, 0.5, 0.25} may be selected to be the
weights. The results are depicted in FIG. 14. In FIG. 14, it is
shown that categories with high social strength include Toys &
Hobbies, Sports Mem, Cards & Fan Shop, Music, Dolls &
Bears, Crafts, Collectibles, and Clothing Shoes & Accessories,
which tracks observations based on collected data. Travel,
Specialty Services, and Computers & Networking are categories
without a strong social community. Travel and Specialty Services
are relatively immature with small numbers of transactions. The
Computer & Networking is a typical retailer mode, in which
there are not many social interactions involved.
[0064] In addition to analyzing the social commerce network using
the degree distribution and the social strength measures, the local
structure is also analyzed. The local structure can be revealed by
isolating recurring small connected sub-networks in the network,
referred to as motifs. Motifs are those patterns for which the
probability P of appearing in a randomized network an equal or
greater number of times than in the real network is lower than a
cutoff value (here, for example, P=0.01). In other words, motifs
are small connected sub-networks that occur in significantly higher
frequencies than would be expected in random networks with similar
node characteristics.
[0065] FIG. 15 depicts examples of motifs, according to some
embodiments. Motif detection may provide some insights to the local
structure of complex networks, while most complex network
measurements can only give global knowledge of the network. Common
motifs may imply common network functionality and reveal structural
evolution principles of complex networks.
[0066] One measure is that of significance profiles of the
sub-networks. The significance Z.sub.i of a sub-network i is
computed by comparing its appearances in the real network to that
in randomly generated networks with the same degree distribution.
For example,
Z.sub.i=(Nreal.sub.i<Nrand.sub.i>)/std(Nrand.sub.i)
where Nreal.sub.i is the count of appearances of the sub-network i
in the real network, and <Nrand.sub.i> and std(Nrand) are the
mean and standard deviation of its appearances in the randomly
generated networks. The significance profile (SP) is the normalized
vector of Z.sub.i:
SP.sub.i=Z.sub.i/(.SIGMA.Z.sub.j.sup.2).sup.1/2
[0067] In some embodiments, triad significance profiles (TSP) are
computed. The TSPs are the significances of the 13 possible
directed connected triads (e.g., three-node motifs). To compute the
SP for a n-node sub-network, all n-node sub-networks in a network
are numerated, which is computationally intensive with time
complexity of O(|V|.sup.n). In one instance, FANMODError! Reference
source not found., a fast network motif detection tool was used.
The TSP may only be calculated for relatively small categories,
including Baby, Dolls & Bears, DVD & Movies, Stamps, and
Crafts and Consumer Electronics. The results are depicted in FIG.
16.
[0068] Based on FIG. 16, the categories except Baby show similar
patterns, in which triad 7, 9, and 10 are strong. The triad 7,
feed-forward loop, is the most significant one in the social
commerce network. The sub-network of Baby has very strong triad 10,
but insignificant triad 7, which might due to sampling errors.
[0069] For complex networks, significant connected sub-networks,
such as triads 7, 9, and 10 in the social commerce networks, are
termed "network motifs." Motifs can be thought of as essential
building blocks of complex networks. The local structure of a
network may be shown by presenting its motifs. Moreover, semantic
information into the motifs may be incorporated into the motifs.
Vertices can be colored by corresponding users' feedback scores.
For example, triads with the same linking structure, but different
coloring of vertices are considered different. In the depicted
example, the continuous feedback score was divided into nine ranges
and colored the vertices accordingly. The 4-node colored motifs are
searched in Stamps and Antiques categories. The top ten motifs of
both categories are listed in FIG. 18.
[0070] FIG. 17 illustrates a method 1700 for the conducting motif
analysis according to an example embodiment. The method 1700 may be
performed by the motif subsystem 112 (FIG. 1).
[0071] User interaction data associated for users for a time period
in a social network (e.g., a social commerce network) is accessed
at block 1702. Network analysis is performed on the user
interaction data at block 1704. The user interaction data may be
collected over more than one time period. The data from the time
periods may be compared.
[0072] Example users within the social commerce network are
selected at block 1706. The example users may be associated with
reputation information (e.g., user feedback).
[0073] A motif for the example users for the time period is
generated in accordance with the performing of the network analysis
at block 1708. A node of the motif may be associated with an
example user. The motif may define an expected relationship between
a number of example users in the social commerce network. For
example, a four node motif may be generated.
[0074] The node of the example users may be distinguished in
accordance with the reputation information of a respective example
user at block 1710. For example, the node of the example users may
be colored in accordance with the reputation information.
[0075] At decision block 1712, a determination may be made whether
to apply a texture to the motif. If a determination is made to
apply a texture, additional user data may be accessed at block 1714
and the texture may be applied to one or more connected lines of
the motif in accordance with the additional user data at block
1716. For example, the additional user data may include interaction
frequency data and/or transaction financial data associated with
the users. If a determination is made not to apply the texture at
decision block 1712 or upon completion of the operations at block
1716, the method 1700 may proceed to the block 1718.
[0076] The motif with the distinguished nodes may be utilized for
analysis of the social commerce network at block 1718. For example,
the motif with the distinguished nodes may be provided for
presentation. In some instances, the method 1700 may be repeated
using user interaction data collected over other time periods.
[0077] FIG. 18 is an example motif display 1800 according to an
example embodiment. The motif display 1800 is an example
representation of four node motifs from two categories of a social
commerce network. However, motifs may be made for other categories
of the social commerce network or the entire social commerce
network. Motifs containing a different number of nodes may also be
used.
[0078] The motif display 1800 includes a number of motifs 1826-1844
for a first category 1802 and a number of motifs 1846 for a second
category 1864. The motifs in two categories are very different. In
stamps category, the motifs suggest that there are second tier
sellers, who bought from users with higher reputation and sold to
users will lower reputation. This structure is not presented in the
motifs of antiques category.
[0079] A distinguishing legend 1806 may include a series of
distinguishing levels 1880-1824 that reflect different reputation
information associated with users of the motifs 1826-1864. For
example, the nodes of the motif 1826 includes a first node with a
distinguishing level 1818, a second node with a distinguishing
level 1820, a third node with a distinguishing level 1822, and a
fourth node with a distinguishing level 1824.
[0080] In addition to studying motifs to analyze the local
structure of the social commerce network, preferential connections
between users may be analyzed. People with many connections tend to
know others with many connections. For the social commerce network,
the users' tendency to deal with others with respect of their
feedback scores is analyzed. In other words, "Do users with high
feedback scores tend to deal with users that also have high
feedback scores?" Assortative mixing may be used to show the extent
to which nodes connect preferentially to other nodes with similar
characteristics. Social commerce networks often show positive
assortativity.
[0081] FIG. 19 illustrates a method 1900 for differentiated
plotting analysis according to an example embodiment. The method
1900 may be performed by the preferential connections subsystem 114
(FIG. 1).
[0082] Reputation information associated with a number of
initiating users and a number of responding users in a social
commerce network for a time period is accessed at block 1902.
[0083] Interaction frequency data associated with the initiating
users and the responding users for the time period is accessed at
block 1904. An aggregated correlation between the initiating users
and the responding users is plotted in accordance with the
reputation information at block 1906.
[0084] The plotting of the aggregated correlation is differentiated
in accordance with the interaction frequency data at block 1908. In
an example embodiment, the differentiated plotting may be used to
determine users' tendency to interact with others with respect of
their reputation information (e.g., feedback scores). For example,
the differentiated plotting may help identify whether users with
high reputation information (e.g., high feedback scores) tend to
interact with other users that also have high reputation
information. Assortative mixing may be used to show the extent to
which nodes (e.g., users) connect preferentially to other nodes
with similar characteristics.
[0085] With reference to FIG. 20, the aggregated correlation
between sellers' feedback scores and buyers' feedback scores was
plotted. In an Auroral diagram, the x-axis denotes sellers'
feedback scores, and the y-axis denotes buyers' feedback scores.
Both axes are logarithmically binned. The color of each block is
determined by the number of transactions happened between pairs of
users with corresponding feedback scores.
[0086] Referring again back to FIG. 19, the differentiated plotting
of the aggregated correlation is utilized at block 1910. For
example, the differentiated plotting of the aggregated correlation
may be provided for presentation.
[0087] FIG. 20 depicts example Auroral diagrams. Diagram 2000 is
based on all transactions in the social commerce network. Diagram
2002 is based on transactions in the crafts category. Diagram 2004
is based on transaction in the collectible category. Diagram 2006
is based on transactions in the computers and networking category.
Diagram 2000 that shows that most transactions happened between
buyers with feedback values within [0, 300] and sellers with
feedback values within [150, 5000]. Sellers with feedback close to
0 only made a few sales, while buyers with feedback close to 0 made
a lot of purchases. This suggests that new buyers are much easier
to "survive" in the social commerce network than new sellers.
Buyers with negative feedbacks made very few transactions
indicating their vulnerability in this network. Comparing crafts
and collectibles, collectibles are more welcoming to new buyers.
Computers and networking category is very different. The core is
squeezed. Sellers with a wider range of feedback scores do well in
this category.
Modules, Components and Logic
[0088] Certain embodiments of subsystems as described herein may be
implemented as logic or a number of modules, components, or
mechanisms. A module, logic, component, or mechanism (herein after
collectively referred to as a "module") may be a tangible unit
capable of performing certain operations and is configured or
arranged in a certain manner. In example embodiments, one or more
computer systems (e.g., a standalone, client or server computer
system) or one or more components of a computer system (e.g., a
processor or a group of processors) may be configured by software
(e.g., an application or application portion) as a "module" that
operates to perform certain operations as described herein.
[0089] In various embodiments, a "module" may be implemented
mechanically or electronically. For example, a module may comprise
dedicated circuitry or logic that is permanently configured (e.g.,
within a special-purpose processor) to perform certain operations.
A module may also comprise programmable logic or circuitry (e.g.,
as encompassed within a general-purpose processor or other
programmable processor) that is temporarily configured by software
to perform certain operations. It will be appreciated that the
decision to implement a module mechanically, in the dedicated and
permanently configured circuitry, or in temporarily configured
circuitry (e.g., configured by software) may be driven by cost and
time considerations.
[0090] Accordingly, the term "module" should be understood to
encompass a tangible entity, be that an entity that is physically
constructed, permanently configured (e.g., hardwired) or
temporarily configured (e.g., programmed) to operate in a certain
manner and/or to perform certain operations described herein.
Considering embodiments in which modules or components are
temporarily configured (e.g., programmed), each of the modules or
components need not be configured or instantiated at any one
instance in time. For example, where the modules or components
comprise a general-purpose processor configured using software, the
general-purpose processor may be configured as respective different
modules at different times. Software may accordingly configure the
processor to constitute a particular module at one instance of time
and to constitute a different module at a different instance of
time.
[0091] Modules can provide information to, and receive information
from, other modules. Accordingly, the described modules may be
regarded as being communicatively coupled. Where multiple of such
modules exist contemporaneously, communications may be achieved
through signal transmission (e.g., over appropriate circuits and
buses) that connect the modules. In embodiments in which multiple
modules are configured or instantiated at different times,
communications between such modules may be achieved, for example,
through the storage and retrieval of information in memory
structures to which the multiple modules have access. For example,
a one module may perform an operation, and store the output of that
operation in a memory device to which it is communicatively
coupled. A further module may then, at a later time, access the
memory device to retrieve and process the stored output. Modules
may also initiate communications with input or output devices, and
can operate on a resource (e.g., a collection of information).
Electronic Apparatus and System
[0092] Example embodiments may be implemented in digital electronic
circuitry, or in computer hardware, firmware, software, or in
combinations of them. Example embodiments may be implemented using
a computer program product, e.g. a computer program tangibly
embodied in an information carrier, e.g., in a machine-readable
medium for execution by, or to control the operation of, data
processing apparatus, e.g., a programmable processor, a computer,
or multiple computers.
[0093] A computer program can be written in any form of programming
language, including compiled or interpreted languages, and it can
be deployed in any form, including as a stand-alone program or as a
module, component, subroutine, or other unit suitable for use in a
computing environment. A computer program can be deployed to be
executed on one computer or on multiple computers at one site or
distributed across multiple sites and interconnected by a
communication network.
[0094] In example embodiments, operations may be performed by one
or more programmable processors executing a computer program to
perform functions by operating on input data and generating output.
Method operations can also be performed by, and apparatus of
example embodiments may be implemented as, special purpose logic
circuitry, e.g., an FPGA (field programmable gate array) or an ASIC
(application-specific integrated circuit).
[0095] The computing system can include clients and servers. A
client and server are generally remote from each other and
typically interact through a communication network. The
relationship of client and server arises by virtue of computer
programs running on the respective computers and having a
client-server relationship to each other. In embodiments deploying
a programmable computing system, it will be appreciated that that
both hardware and software architectures require consideration.
Specifically, it will be appreciated that the choice of whether to
implement certain functionality in permanently configured hardware
(e.g., an ASIC), in temporarily configured hardware (e.g., a
combination of software and a programmable processor), or a
combination permanently and temporarily configured hardware may be
a design choice. Below are set out hardware (e.g., machine) and
software architectures that may be deployed, in various example
embodiments.
Example Machine Architecture and Machine-Readable Medium
[0096] An example embodiment extends to a machine in the example
form of a computer system 2100 within which instructions, for
causing the machine to perform any one or more of the methodologies
discussed herein, may be executed. In alternative embodiments, the
machine operates as a standalone device or may be connected (e.g.,
networked) to other machines. In a networked deployment, the
machine may operate in the capacity of a server or a client machine
in server-client network environment, or as a peer machine in a
peer-to-peer (or distributed) network environment. The machine may
be a personal computer (PC), a tablet PC, a set-top box (STB), a
Personal Digital Assistant (PDA), a cellular telephone, a web
appliance, a network router, switch or bridge, or any machine
capable of executing instructions (sequential or otherwise) that
specify actions to be taken by that machine. Further, while only a
single machine is illustrated, the term "machine" shall also be
taken to include any collection of machines that individually or
jointly execute a set (or multiple sets) of instructions to perform
any one or more of the methodologies discussed herein.
[0097] The example computer system 2100 includes a processor 2102
(e.g., a central processing unit (CPU), a graphics processing unit
(GPU) or both), a main memory 2104 and a static memory 2106, which
communicate with each other via a bus 2108. The computer system
2100 may further include a video display unit 2110 (e.g., a liquid
crystal display (LCD) or a cathode ray tube (CRT)). The computer
system 2100 also includes an alphanumeric input device 2112 (e.g.,
a keyboard), a user interface (UI) navigation device 2114 (e.g., a
mouse), a disk drive unit 2116, a signal generation device 2118
(e.g., a speaker) and a network interface device 2120 to
communicate via a communications network 2126.
Machine-Readable Medium
[0098] The disk drive unit 2116 includes a non-transitory
machine-readable medium 2122 on which is stored one or more sets of
instructions and data structures (e.g., software 2124) embodying or
utilized by any one or more of the methodologies or functions
described herein. The software 2124 may also reside, completely or
at least partially, within the main memory 2104 and/or within the
processor 2102 during execution thereof by the computer system
2100, the main memory 2104 and the processor 2102 also constituting
machine-readable media.
[0099] While the non-transitory machine-readable medium 2122 is
shown in an example embodiment to be a single medium, the term
"non-transitory machine-readable medium" may include a single
medium or multiple media (e.g., a centralized or distributed
database, and/or associated caches and servers) that store the one
or more instructions. The term "non-transitory machine-readable
medium" shall also be taken to include any tangible medium that is
capable of storing, encoding or carrying instructions for execution
by the machine and that cause the machine to perform any one or
more of the methodologies of the present invention, or that is
capable of storing, encoding or carrying data structures utilized
by or associated with such instructions. The term "non-transitory
machine-readable medium" shall accordingly be taken to include, but
not be limited to, solid-state memories, and optical and magnetic
media. Specific examples of machine-readable media include
non-volatile memory, including by way of example semiconductor
memory devices, e.g., EPROM, EEPROM, and flash memory devices;
magnetic disks such as internal hard disks and removable disks;
magneto-optical disks; and CD-ROM and DVD-ROM disks.
Transmission Medium
[0100] The software 2124 may further be transmitted or received
over the communications network 2126 using a transmission medium
via the network interface device 2120 utilizing any one of a number
of well-known transfer protocols (e.g., HTTP). Examples of
communication networks include a local area network ("LAN"), a wide
area network ("WAN"), the Internet, mobile telephone networks,
Plain Old Telephone (POTS) networks, and wireless data networks
(e.g., WiFi and WiMax networks) The term "transmission medium"
shall be taken to include any intangible medium that is capable of
storing, encoding or carrying instructions for execution by the
machine, and includes digital or analog communications signals or
other intangible medium to facilitate communication of such
software.
Example Three-Tier Software Architecture
[0101] In some embodiments, the described methods may be
implemented using one a distributed or non-distributed software
application designed under a three-tier architecture paradigm.
Under this paradigm, various parts of computer code (or software)
that instantiate or configure components or modules may be
categorized as belonging to one or more of these three tiers. Some
embodiments may include a first tier as an interface (e.g., an
interface tier). Further, a second tier may be a logic (or
application) tier that performs application processing of data
inputted through the interface level. The logic tier may
communicate the results of such processing to the interface tier,
and/or to a backend, or storage tier. The processing performed by
the logic tier may relate to certain rules, or processes that
govern the software as a whole. A third, storage tier, may be a
persistent storage medium, or a non-persistent storage medium. In
some cases, one or more of these tiers may be collapsed into
another, resulting in a two-tier architecture, or even a one-tier
architecture. For example, the interface and logic tiers may be
consolidated, or the logic and storage tiers may be consolidated,
as in the case of a software application with an embedded database.
The three-tier architecture may be implemented using one
technology, or, a variety of technologies. The example three-tier
architecture, and the technologies through which it is implemented,
may be realized on one or more computer systems operating, for
example, as a standalone system, or organized in a server-client,
peer-to-peer, distributed or so some other suitable configuration.
Further, these three tiers may be distributed between more than one
computer systems as various components.
Components
[0102] Example embodiments may include the above described tiers,
and processes or operations about constituting these tiers may be
implemented as components. Common too many of these components is
the ability to generate, use, and manipulate data. The components,
and the functionality associated with each, may form part of
standalone, client, server, or peer computer systems. The various
components may be implemented by a computer system on an as-needed
basis. These components may include software written in an
object-oriented computer language such that a component oriented,
or object-oriented programming technique can be implemented using a
Visual Component Library (VCL), Component Library for Cross
Platform (CLX), Java Beans (JB), Java Enterprise Beans (EJB),
Component Object Model (COM), Distributed Component Object Model
(DCOM), or other suitable technique.
[0103] Software for these components may further enable
communicative coupling to other components (e.g., via various
Application Programming interfaces (APIs)), and may be compiled
into one complete server, client, and/or peer software application.
Further, these APIs may be able to communicate through various
distributed programming protocols as distributed computing
components.
Distributed Computing Components and Protocols
[0104] Some example embodiments may include remote procedure calls
being used to implement one or more of the above described
components across a distributed programming environment as
distributed computing components. For example, an interface
component (e.g., an interface tier) may form part of a first
computer system that is remotely located from a second computer
system containing a logic component (e.g., a logic tier). These
first and second computer systems may be configured in a
standalone, server-client, peer-to-peer, or some other suitable
configuration. Software for the components may be written using the
above described object-oriented programming techniques, and can be
written in the same programming language, or a different
programming language. Various protocols may be implemented to
enable these various components to communicate regardless of the
programming language used to write these components. For example, a
component written in C++ may be able to communicate with another
component written in the Java programming language through
utilizing a distributed computing protocol such as a Common Object
Request Broker Architecture (CORBA), a Simple Object Access
Protocol (SOAP), or some other suitable protocol. Some embodiments
may include the use of one or more of these protocols with the
various protocols outlined in the Open Systems Interconnection
(OSI) model, or Transmission Control Protocol/Internet Protocol
(TCP/IP) protocol stack model for defining the protocols used by a
network to transmit data.
A System of Transmission Between a Server and Client
[0105] Example embodiments may use the OSI model or TCP/IP protocol
stack model for defining the protocols used by a network to
transmit data. In applying these models, a system of data
transmission between a server and client, or between peer computer
systems may for example include five layers comprising: an
application layer, a transport layer, a network layer, a data link
layer, and a physical layer. In the case of software, for
instantiating or configuring components, having a three tier
architecture, the various tiers (e.g., the interface, logic, and
storage tiers) reside on the application layer of the TCP/IP
protocol stack. In an example implementation using the TCP/IP
protocol stack model, data from an application residing at the
application layer is loaded into the data load field of a TCP
segment residing at the transport layer. This TCP segment also
contains port information for a recipient software application
residing remotely. This TCP segment is loaded into the data load
field of an IP datagram residing at the network layer. Next, this
IP datagram is loaded into a frame residing at the data link layer.
This frame is then encoded at the physical layer, and the data
transmitted over a network such as an internet, Local Area Network
(LAN), Wide Area Network (WAN), or some other suitable network. In
some cases, internet refers to a network of networks. These
networks may use a variety of protocols for the exchange of data,
including the aforementioned TCP/IP, and additionally ATM, SNA,
SDI, or some other suitable protocol. These networks may be
organized within a variety of topologies (e.g., a star topology),
or structures.
[0106] Although an embodiment has been described with reference to
specific example embodiments, it will be evident that various
modifications and changes may be made to these embodiments without
departing from the broader spirit and scope of the invention.
Accordingly, the specification and drawings are to be regarded in
an illustrative rather than a restrictive sense. The accompanying
drawings that form a part hereof, show by way of illustration, and
not of limitation, specific embodiments in which the subject matter
may be practiced. The embodiments illustrated are described in
sufficient detail to enable those skilled in the art to practice
the teachings disclosed herein. Other embodiments may be utilized
and derived therefrom, such that structural and logical
substitutions and changes may be made without departing from the
scope of this disclosure. This Detailed Description, therefore, is
not to be taken in a limiting sense, and the scope of various
embodiments is defined only by the appended claims, along with the
full range of equivalents to which such claims are entitled.
[0107] Such embodiments of the inventive subject matter may be
referred to herein, individually and/or collectively, by the term
"invention" merely for convenience and without intending to
voluntarily limit the scope of this application to any single
invention or inventive concept if more than one is in fact
disclosed. Thus, although specific embodiments have been
illustrated and described herein, it should be appreciated that any
arrangement calculated to achieve the same purpose may be
substituted for the specific embodiments shown. This disclosure is
intended to cover any and all adaptations or variations of various
embodiments. Combinations of the above embodiments, and other
embodiments not specifically described herein, will be apparent to
those of skill in the art upon reviewing the above description.
* * * * *