JP5089631B2 - Relevance analysis system and relevance analysis method - Google Patents

Description

  The present invention relates to a relationship analysis system and a relationship analysis method for analyzing relationships between users.

  Opinions of various users are transmitted on the Internet by information transmission media such as blogs and SNSs. Based on such user opinions, calculation of the strength of relevance between users, extraction of a user community using a clustering technique, and the like are performed (for example, see Patent Document 1 and Patent Document 2).

JP 2007-94552 A JP 2008-59442 A

  However, when analyzing the relevance between users based on the user information transmitted by the information transmission medium, which has been known in the past, the relevance between users with similar philosophy or similar relevance between users is discovered Although there is a problem, there is a problem that it is impossible to discover the relevance between users who are strongly connected in daily life or the relevance between users who are taking similar actions. Moreover, since the user information transmitted by the information transmission medium includes a lot of information that becomes noise, a large amount of information is required to improve the analysis accuracy. For this reason, in order to obtain a sufficient analysis result, a large amount of information must be acquired and analyzed.

  The present invention has been made in consideration of such circumstances, and it is possible to reduce the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions with less information. It is an object of the present invention to provide a relevance analysis system and a relevance analysis method that can be analyzed based on the above.

The present invention generates a relevance information indicating relevance between the plurality of users based on the schedule information management unit that acquires the schedule information of the plurality of users and the schedule information acquired by the schedule information management unit. A schedule information analysis unit; and an image generation unit that generates an image indicating the relationship between the plurality of users based on the relevance information generated by the schedule information analysis unit . The schedule information analysis unit generates information indicating the number of times that the user and a user different from the user execute the same schedule for each combination of the user and a user different from the user, The schedule information analysis unit generates a different user from the user who executed the same schedule as the user. That it contained the information indicating the an association analysis system according to claim.

  According to this configuration, the schedule information management unit acquires schedule information of a plurality of users. A schedule information analysis part produces | generates the relevance information which shows the relevance between several users based on the schedule information which the schedule information management part acquired. An image generation part produces | generates the image which shows the relationship between several users based on the relevance information which the schedule information analysis part produced | generated. That is, only schedule information indicating a user's daily life and user behavior is acquired, and the relevance between a plurality of users is analyzed based on the acquired schedule information. Thereby, it is possible to analyze the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions based on less information.

Further, the present invention provides a schedule information management step in which a schedule information management unit acquires schedule information of a plurality of users, and a schedule information analysis unit based on the schedule information acquired in the schedule information management step. A schedule information analysis step for generating relevance information indicating the relevance between the users, and an image generation unit that determines the relevance between the plurality of users based on the relevance information generated in the schedule information analysis step. seen containing an image generating step of generating an image showing the the relationship information, in the schedule information analyzing step, information indicating the number of times and different user from the user and the user performs the same schedule, Information generated for each combination of the user and the user different from the user It said schedule information analysis generated in step, the user with the user executing the same schedule and the related analysis method characterized by including the information indicating the number of different users.

  According to the present invention, only schedule information indicating a user's daily life and user behavior is acquired, and the relevance between a plurality of users is analyzed based on the acquired schedule information. Thereby, it is possible to analyze the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions based on less information.

It is a block diagram which shows the structure of the relationship analysis system in one Embodiment of this invention. It is the figure which showed the data structure of the schedule information table in this embodiment. It is the figure which showed the data structure of the link score table in this embodiment. It is the figure which showed the data structure of the keyword score table in this embodiment. It is the figure which showed the data structure of the user information table in this embodiment. It is the flowchart which showed the procedure in which the schedule information management server in this embodiment updates a schedule update information table. It is the flowchart which showed the procedure in which the schedule information analysis system in this embodiment updates a link score table and a keyword score table. It is the flowchart which showed the procedure in which the graph production | generation system in this embodiment produces | generates the source file for producing | generating an image. It is the figure which showed the data structure of the connection matrix in this embodiment. It is the figure which showed the user relevance graph which the image generation function part in this embodiment produced | generated.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the relevance analysis system in this embodiment. In the illustrated example, the relevance analysis system includes a scheduler 101, a schedule information management server 102 (schedule information management unit), a schedule information database 103, a schedule information analysis system 104 (schedule information analysis unit), and a user relevance. A database 105, a graph generation system 106 (image generation unit), and an image generation function unit 107 (image generation unit) are provided.

  The scheduler 101 stores user schedule information. For example, schedule information input by the user is stored. The schedule information management server 102 periodically acquires schedule information stored in the scheduler 101. The schedule information database 103 stores schedule information acquired by the schedule information management server 103.

The schedule information analysis system 104 reads the schedule information stored in the schedule information database 103 and analyzes the relationship between users. The user relevance database 105 stores the results (relevance information) analyzed by the schedule information analysis system 104. The graph generation system 106 reads the relevance information stored in the user relevance database 105 and generates information for generating an image indicating the relevance between users. The image generation function unit 107 generates an image indicating the relationship between users based on the information generated by the graph generation system 106 for generating an image indicating the relationship between users.

In addition, the user terminal 108, the scheduler 101 included in the relevance analysis system, and the graph generation system 106 are connected via a communicable network and can transmit and receive information to and from each other. Thereby, the user terminal 108 can input the user's schedule information to the scheduler 101. Further, the user terminal 108 can request the graph generation system 106 to transmit an image indicating the relationship between users. Each unit included in the relevance analysis system is realized by one or a plurality of server devices.

  Next, information stored in the schedule information database 103 will be described. FIG. 2 is a diagram showing a data structure of the schedule information table 201 stored in the schedule information database 103.

  The schedule information table 201 has attributes of ID 202, user ID 203, date 204, start time 205, end time 206, event content 207, and participant list 208.

  The ID 202 is an identifier given for each entry in the schedule information table 201. The user ID 203 is an identifier uniquely given to each user over the entire system. The date 204 is the date when the user specified by the user ID 203 executes the event stored in the event content 207. The start time 205 is a time when the user specified by the user ID 203 starts the event content 207. The end time 206 is a time when the user specified by the user ID 203 ends the event content 207. The event content 207 is the subject of the schedule. The participant list 208 is a list of user IDs that specify users who have participated in the event content 207.

  For example, in the entry identified by ID 202 being “1”, the user identified by user ID 203 “3020” made a meeting from 13:00 to 14:00 on October 10, 2008, and the user was assigned to this meeting. This indicates that the user whose ID 203 is specified by “3030, 3040, 3041, 3042” participates. Other entries are as shown.

  Next, information stored in the user relevance database 105 will be described. FIG. 3 is a diagram showing a data structure of the link score table 301 stored in the user relevance database 105.

  The link score table 301 has attributes of ID 302, user ID 303, user name 304, link hash 305, and known user array 306.

  The ID 302 is an identifier given for each entry in the link score table 301. The user ID 303 is information similar to the user ID 203 in the schedule information table 201, and is an identifier that is uniquely given to each user throughout the system. The user name 304 is the name of the user specified by the user ID 303. The link hash 305 is a hash value obtained by hashing a combination of the user ID 303 of the schedule sharing user for the user name 304 and the number of times of sharing. The key of this hash value is the user ID 303, and the value is the number of sharing. The known user array 306 is an array of a list of user IDs 303 of users whose user names 304 have shared a schedule.

  In this embodiment, the hash value and array representation method uses the data structure representation of Ruby language.

  For example, in an entry identified by ID 302 of “1”, the user name 340 of the user identified by user ID 303 “3020” is “yamada.ichirou” and the link hash 305 is “{“ 6080 ”: 10, “6090”: 15, “3000”: 3, “3030”: 30} ”and the known user array 306 is“ [6080,6090,3000,3030] ”. Other entries are as shown.

  FIG. 4 is a diagram illustrating a data structure of the keyword score table 401 stored in the user relevance database 105. The keyword score table 401 has attributes of ID 402, user ID 403, sharer user ID 404, and keyword hash 405.

  The ID 402 is an identifier given for each entry in the keyword score table 401. The user ID 403 is the same information as the user ID 203 of the schedule information table 201 and the user ID 303 of the link score table 301, and is an identifier that is uniquely given to each user over the entire system. The sharer user ID 404 is a user ID 403 of a schedule sharing user who shares a schedule with the user ID 403. The keyword hash 405 is a combination of a word obtained by the morpheme analysis and the number of appearances of the morpheme analysis performed on the event content 207 in which the user specified by the user ID 403 and the user specified by the sharer user ID 404 participate. Is a hash value obtained by hashing The key of this hash value is a word obtained by morphological analysis, and the value is the number of appearances.

  For example, in an entry identified by ID 402 being “1”, user ID 403 is “3020”, sharer user ID 404 is “3030”, and keyword hash 405 is “{“ meeting ”: 12,“ meeting ”: 8, “Business trip”: 20, “Director Tanaka”: 20} ”. Other entries are as shown.

  FIG. 5 is a diagram showing a data structure of the user information table 501 stored in the user relevance database 105. The user information table 501 has attributes of ID 502, user ID 503, group ID 504, user name 505, and belonging group name 506.

  The ID 502 is an identifier given for each entry in the user information table 501. The user ID 503 is the same information as the user ID 203 of the schedule information table 201, the user ID 303 of the link score table 301, and the user ID 403 of the keyword score table 401, and is an identifier that is uniquely given to each user over the entire system. is there. The group ID 504 is an identifier uniquely given to each group to which the user belongs throughout the entire system. The user name 505 is the same information as the user name 304 in the link score table 301, and is the name of the user specified by the user ID 503. The belonging group name 506 is the name of the group specified by the group ID 504.

  For example, an entry identified by ID 502 of “1” has a user ID 503 of “3020”, a group ID 504 of “999”, a user name 505 of “yamada.ichirou”, and an affiliated group 506 of “soumu”. Is. Other entries are as shown.

  In the user information table 501, information about users who use the relevance analysis system is defined in advance.

Next, the operation of the relevance analysis system will be described.
[Update processing of schedule information database 103]
First, a method for updating the schedule information table 201 stored in the schedule information database 103 will be described. The schedule information management server 103 updates the schedule update information table 201. FIG. 6 is a flowchart showing a procedure for updating the schedule update information table 201 by the schedule information management server 103.

  (Step S101) The schedule information management server 103 reads schedule information stored in the scheduler 101. Thereafter, the process proceeds to step S102. The schedule information includes a user ID 203, a date 204, a start time 205, an end time 206, an event content 207, and a participant list 208.

  For example, the scheduler 101 provides an API (Application Program Interface), and the schedule information management server 103 reads schedule information from the scheduler 101 using this API.

  (Step S102) The schedule information management server 103 compares the schedule information acquired in step S101 with the information stored in the schedule update information table 201 of the schedule information database 103. Thereafter, the process proceeds to step S103.

  (Step S103) If the schedule update information table 201 determines that the schedule information acquired in step S101 is not stored in the schedule update information table 201 as a result of the comparison in step S102, the process proceeds to step S104. The process proceeds to S105.

  (Step S104) The schedule information management server 103 stores the schedule information acquired in step S101 in the schedule update information table 201. Thereafter, the process proceeds to step S105.

  (Step S105) The schedule information management server 103 determines whether or not all schedule information stored in the scheduler 101 has been acquired in step S101. If the schedule information management server 103 determines in step S101 that all the schedule information stored in the scheduler 101 has been acquired, the process ends, otherwise the process returns to step S101.

  The schedule update information table 201 can store the schedule information stored in the scheduler 101 by the operations from step S101 to step S105.

[Update processing of user relevance database 105]
Next, a method for updating the link score table 301 and the keyword score table 401 of the user relevance database 105 will be described. The schedule information analysis system 104 updates the link score table 301 and the keyword score table 401. FIG. 7 is a flowchart showing a procedure in which the schedule information analysis system 104 updates the link score table 301 and the keyword score table 401.

  (Step S201) The schedule information analysis system 104 reads all the schedule information stored in the schedule information table 201 of the schedule information database 103. Thereafter, the process proceeds to step S202.

  (Step S202) The schedule information analysis system 104 determines one specific user ID 203 among the user IDs 203 included in the schedule information read in step S201. Subsequently, the schedule information analysis system 104 schedules the specific user ID 203 for each other user ID from the combination of the user ID 203 and the participant list 208 included in the schedule information acquired in step S201. Calculate the number of shares. Thereafter, the process proceeds to step S203.

  For example, based on the schedule information table 201 shown in FIG. 2, the result of the schedule information analysis system 104 calculating the number of times the specific user ID “3020” shares the schedule for each other user ID is It is as follows.

  The user ID “3020” shares the schedule once with the user ID “3030”, shares the schedule once with the user ID “3040”, and shares the schedule once with the user ID “3041”. The user ID “3042” is shared once, the user ID “2000” is shared once, and the user ID “2010” is shared once.

  (Step S203) The schedule information analysis system 104 generates a link hash 305 and a known user array 306 for a specific user ID 203 based on the result calculated in Step S202. Thereafter, the process proceeds to step S204.

  (Step S204) The schedule information analysis system 104 reads the user name 505 associated with the specific user ID 203 from the user information table 501 of the user relevance database 105. Subsequently, the schedule information analysis system 104 associates the specific user ID 203, the user name 505 read from the user information table 501, the link hash 305 generated in step S203, and the known user array 306, and associates the user relevance. It is stored in the link score table 301 of the database 105. Thereafter, the process proceeds to step S205.

  (Step S205) The schedule information analysis system 104 analyzes the morphological analysis of the event content 207 shared by the specific user ID 203 and another user ID (sharer user ID 404) based on the schedule information read in step S201. Is performed for each other user ID to extract a word included in the event content 207 and further calculate the number of appearances of the extracted word. Thereafter, the process proceeds to step S206.

  (Step S206) The schedule information analysis system 104 determines the keyword from the word included in the event content 207 specified by the specific user ID 203 and other user ID extracted in Step S205, and the number of times this word appears. A hash 405 is generated. Subsequently, the schedule information analysis system 104 associates a specific user ID 203 (user ID 403) with another user ID (sharer user ID 404) and the generated keyword hash 405, and generates a keyword score table of the user relevance database 105. 401 is stored. Thereafter, the process proceeds to step S207.

  (Step S207) The schedule information analysis system 104 determines whether or not all user IDs 203 included in the schedule information read in step S201 have been set as specific user IDs 203 in step S202. That is, it is determined whether or not the sharer user ID 404 and the keyword hash 405 have been generated for all the user IDs 203 included in the schedule information read in step S201. If the schedule information analysis system 104 determines that all the user IDs 203 included in the schedule information read in step S201 are the specific user IDs 203 in step S202, the process ends. Otherwise, the process returns to step S201.

  As a result, a sharer user ID 404 and a generated keyword hash 405 related to the user ID 203 can be generated.

[Generate image]
Next, a method for generating an image showing a user relevance graph by the graph generation system 106 and the image generation function unit 107 will be described. This user relationship graph is a graph showing the relationship between users who are strongly connected in daily life and the relationship between users who are taking similar actions.

  FIG. 8 is a flowchart showing a procedure by which the graph generation system 106 generates a source file for the image generation function unit 107 to generate an image.

  (Step S301) The graph generation system 106 reads all information stored in the link score table 301 of the user relevance database 105. Thereafter, the process proceeds to step S302.

  (Step S302) The graph generation system 106 generates a connection matrix based on the user ID 303 and the user ID included in the link hash 305 among the information read out in step S301. The connection matrix is a matrix indicating the number of times the user specified by the user ID 303 and the user specified by the other user ID 303 share the schedule. Thereafter, the process proceeds to step S303.

  Here, the connection matrix will be described with reference to FIG. FIG. 9 is a diagram showing the data structure of the connection matrix 901. In the illustrated example, the first column and the first row of the connection matrix 901 are the user ID 203. The other rows and columns indicate the number of times that the user specified by the user ID 203 indicated by the first column and the user specified by the user ID 203 indicated by the first row share the schedule. For example, the number of times the user identified by the user ID “2010” and the user identified by the user ID “2020” share the schedule is “40”.

  Note that if the connection matrix is A and the elements of the connection matrix A are Aij (where i ≠ 1, j ≠ 1), only the user specified by the user ID 203 participates in Aii, and other users The user specified by ID 203 is the number of schedules in which the user does not participate. Aij = Aji.

Returning to the description of FIG.
(Step S303) The graph generation system 106 calculates a weight for each user ID 303 based on the known user array 306 included in the information read in step S301. Thereafter, the process proceeds to step S304. In the present embodiment, the weight of the user ID 303 is a value proportional to the size of the known user array 306.

  (Step S304) The graph generation system 106 generates a source file corresponding to the drawing tool used by the image generation function unit 107 based on the connection matrix generated in step S302 and the weight for each user ID 303 generated in step S303. To do. Thereafter, the process ends.

  For example, when the image generation function unit 107 uses Graphviz as a drawing tool, the graph generation system 106 uses the connection matrix generated in step S302 and the weight for each user ID 303 generated in step S303 to display an image in the Graphviz format. Generate source files for

  Based on the source file generated by the graph generation system 106, the image generation function unit 107 shows the relationship between users who are strongly connected in daily life and the relationship between users who are taking similar actions. An image showing the relevance graph is generated.

  FIG. 10 is a diagram showing a user relevance graph generated by the image generation function unit 107. In the illustrated example, the node 1002 represents a user, and the numerical value indicated in the node 1002 is a user ID 203. The size of the node 1002 is a size corresponding to the weight of the user ID 303 calculated in step S303. That is, the size of the node 1002 is proportional to the size of the known user array 306 associated with the user ID 203 in the link score table 301. Thereby, by confirming the size of the node 1002, the number of known users associated with the user ID 203 indicated by the node can be confirmed.

  For example, among the nodes 1002, the node 1002 indicating the user ID “6020” is larger than the other nodes 1002. Thereby, it can be seen that the number of known users associated with the user ID “6020” is larger than the number of known users associated with the other user IDs.

  A link 1003 is a line connecting the nodes 1002 and indicates that the user indicated by the connected node 1002 has shared the schedule. A numerical value attached to the link 1003 indicates the number of times the schedule is shared. The link 1003 and the numerical value attached to the link 1003 are drawn based on the connection matrix calculated in step S302.

  For example, a node 1002 indicating a user ID “6020” and a node 1002 indicating a user ID “6060” are connected by a link 1003. The numerical value assigned to the link 1003 is “8”. Accordingly, it can be seen that the user specified by the user ID “6020” and the user specified by the user ID “6060” share the schedule eight times.

  As described above, according to the present embodiment, only schedule information indicating a user's daily life and user behavior is acquired, and relevance between a plurality of users is analyzed based on the acquired schedule information. Therefore, it is possible to analyze the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions based on less information. Further, based on the analyzed information, it is possible to generate a graph showing the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions. Thereby, it is possible to visually know the relevance between users who are strongly connected in daily life and the relevance between users who are taking similar actions. In addition, it is possible to visually know the connection between everyday users, and to discover the main users in the community. For example, by applying this embodiment to an online scheduler in a company, it can be applied to organization diagnosis and organization management.

  As described above, the embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the scope of the present invention. Is possible.

  In addition, a program for realizing the function of the relevance analysis system is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read and executed by the computer system. May be realized. Here, the “computer system” may include an OS and hardware such as peripheral devices.

  Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used. The “computer-readable recording medium” means a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a CD-ROM, a hard disk built in a computer system, etc. This is a storage device.

  Further, the “computer-readable recording medium” refers to a volatile memory (for example, DRAM (Dynamic) in a computer system serving as a server or a client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. Random Access Memory)) that holds a program for a certain period of time is also included.

  The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.

  DESCRIPTION OF SYMBOLS 101 ... Scheduler, 102 ... Schedule information management server, 103 ... Schedule information database, 104 ... Schedule information analysis system, 105 ... User relevance database, 106 ... Graph generation system, 107 ... Image generation function unit, 108 ... User terminal

Claims (2)

A schedule information management unit for acquiring schedule information of a plurality of users;
Based on the schedule information acquired by the schedule information management unit, a schedule information analysis unit that generates relevance information indicating relevance between the plurality of users;
Based on the relevance information generated by the schedule information analysis unit, an image generation unit that generates an image indicating the relevance between the plurality of users;
Equipped with a,
The relevance information includes information indicating the number of times the user and a user different from the user execute the same schedule for each combination of the user and a user different from the user. The generated information and information indicating the number of users different from the user who executed the same schedule as the user, generated by the schedule information analysis unit.
Relevance analysis system characterized by that. A schedule information management unit for acquiring schedule information of a plurality of users;
A schedule information analyzing step for generating relevance information indicating relevance between the plurality of users based on the schedule information acquired in the schedule information managing step;
An image generating step for generating an image indicating the relevance between the plurality of users based on the relevance information generated in the schedule information analyzing step;
Only including,
The relevance information is information indicating the number of times the user and a user different from the user execute the same schedule in the schedule information analysis step for each combination of the user and a user different from the user. The generated information and information indicating the number of users that are generated in the schedule information analysis step and that are different from the user who executed the same schedule as the user are included.
Relevance analysis method characterized by this.

Images