JP2010020739A - Information management apparatus, method and program for generating, searching and displaying directory reflecting social popularity/interest - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information management apparatus, method and program for generating and displaying a directory reflecting the degree of social interests of a user in a directory type search system. <P>SOLUTION: An information management server 10 manages a plurality of directories, and URLs stored in the directories. The server 10 generates, based on a click data storage 111, a first link with directivity from a URL to a query and a second link with directivity from a query to a selected URL. The server 10 calculates, based on the click data storage 111, a first probability of transition from the URL to the query and a second probability of transition from the query to the selected URL in the generated first link with directivity and a second link with directivity, and calculates a third transition probability based on the first transition probability and the second transition probability. If the third transition probability is larger than a first threshold value and the selected URL exists in the directory, then the server 10 changes the display rank of the selected URL in the directory. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an information management apparatus, method, and program for managing a registration site in a directory system.

  2. Description of the Related Art Conventionally, in the technical field of Web search, a technique for reflecting a user's interest and interest in a link as a URL (Uniform Resource Locator) or a query as a search keyword is known.

For example, Patent Document 1 discloses a technique for managing URLs and automatically categorizing them to create individual and public directories for the links. Patent Document 2 discloses a technique in which a user variably adjusts how much his / her personal interest affects the result of a specific search query.
Special Table 2002-539559 gazette Special Table 2007-531160

  Here, there is a method called a directory type in the Web search. This is a method in which a registered site is arranged in a hierarchical directory reflecting the subject classification, and a searcher accesses the registered site while following the hierarchy.

  On the other hand, in contrast to directory-type fixed classification, in recent cooperative tagging and the like, a method of giving a tag that a user has come up with more freely to an information item is performed. As a result, information items are easily searched according to the degree of social interest in the user's community.

  In the directory-type search method, a case is considered in which the interest of a specific specialized field or the interest of a specific individual is reflected in the directory. In this case, simply reflecting general social popularity and interest in the directory may not sufficiently reflect the interest of a specific specialized field or the interest of a specific individual.

  Therefore, an object of the present invention is to provide an information management apparatus, method, and program for creating and displaying a directory reflecting the degree of social interest of a user in a directory type search method.

  It is another object of the present invention to provide an information management apparatus, method, and program for creating a directory that reflects the interest of a specific specialized field or the interest of a specific individual in a directory-type search method.

  The present invention provides the following solutions.

  (1) A search in which the search query is searched by the search engine in an information management device that manages a URL stored in a directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query Based on the URL and the input query stored in the storage means, the URL clicked by the user and the input query input by the user terminal among the resulting URLs are stored in association with each other. A first link creating means for creating a first directional link to the input query for obtaining the URL as the search result from a URL stored in the directory, and the URL stored in the storage means and the Based on the input query, the input query from the input query constituting the first directed link is Of the URLs of search results searched by the search engine, second link creating means for creating a second directional link to the selected selected URL and the URL constituting the first directional link are stored. Transition probability calculating means for calculating a transition probability from the directory to the selected URL constituting the second directional link, and first threshold value determining means for determining whether or not the transition probability is greater than a first threshold value; If the transition probability is larger than the first threshold as a result of the determination by the first threshold determination means, URL determination means for determining whether the selected URL exists in the directory; and If the selected URL exists in the directory as a result of the determination by the URL determining means, a display order changing procedure for changing the display order of the selected URL in the directory. Information management apparatus, characterized in that it comprises and.

  According to the configuration of (1), the information management device calculates the transition probability from the directory to the selected URL based on the URL and the input query stored in the storage unit, and the transition probability is less than the first threshold value. If it is large and the selected URL exists in the directory, the display order of the selected URL in the directory is changed.

  For this reason, the information management device changes the display order of the selected URL based on the URL and the input query stored in the storage unit reflecting the user's social interest. Therefore, the information management apparatus can reflect the user's social interest in the display order of the selected URL.

  (2) The method according to (1), further comprising: URL adding means for adding the selected URL to the directory when the selected URL does not exist in the directory as a result of the determination by the URL determining means. The information management device described.

  According to the configuration of (2), the information management apparatus adds the selection URL to the directory when the transition probability from the directory to the selection URL is larger than the first threshold and the selection URL does not exist in the directory. .

  For this reason, the information management apparatus adds the selected URL to the directory based on the URL and the input query stored in the storage unit reflecting the user's social interest. Therefore, the information management apparatus can add the selected URL reflecting the social interest of the user to the directory.

  (3) The transition probability calculation means is configured to change the URL constituting the first directional link to the query constituting the first directional link based on the URL and the input query stored in the storage means. Based on the first transition probability calculating means for calculating the first transition probability, and the queries constituting the first directional link and the second directional link based on the URL and the input query stored in the storage means. Based on the second transition probability calculating means for calculating a second transition probability to the selected URL constituting the second directional link, the first directional link based on the first transition probability and the second transition probability. (3) or a third transition probability calculating means for calculating the transition probability from the directory in which the URL constituting the URL to the selected URL constituting the second directional link is included. Information management apparatus according to 2).

  According to the configuration of (3), the information management device includes the first transition probability from the URL configuring the first directional link to the input query configuring the first directional link, the first directional link, and the second A second transition probability from the input query constituting the directed link to the selected URL constituting the second directed link is calculated, and from the directory to the selected URL based on the first transition probability and the second transition probability. The transition probability of is calculated.

  For this reason, the information management apparatus can suitably calculate the transition probability based on the first transition probability and the second transition probability.

  (4) Fourth transition probability calculating means for calculating a fourth transition probability from the directory to an input query constituting the first directional link and the second directional link based on the first transition probability; A second threshold determination means for determining whether or not the fourth transition probability is greater than a second threshold; and, as a result of the determination, if the fourth transition probability is greater than a second threshold, (3) The subdirectory creation means which creates further the subdirectory by which the name of the input query which constitutes the 1st directed link and the 2nd directed link was given to the directory. Information management device.

  According to the configuration of (4), the information management device calculates the fourth transition probability from the directory to the input query that constitutes the first directional link and the second directional link, and the fourth transition probability is the second If it is larger than the threshold value, a subdirectory is created in which the names of the input queries constituting the first directional link and the second directional link are assigned to the directory.

  For this reason, the information management apparatus can create a subdirectory to which the name of the query reflecting the social interest of the user is given.

  (5) Third threshold determination means for determining whether or not the transition probability is greater than a third threshold; and as a result of determination by the third threshold determination means, the transition probability is greater than the third threshold The information management apparatus according to (4), further comprising sub-directory URL adding means for adding the selected URL to the created sub-directory when it is larger.

  According to the configuration of (5), when the transition probability from the directory to the selected URL is larger than the third threshold, the information management apparatus adds the selected URL to the created subdirectory.

  Therefore, the information management device adds the selected URL to the subdirectory based on the URL and the input query stored in the storage unit reflecting the user's social interest. Therefore, the information management apparatus can add a selection URL reflecting the social interest of the user to the subdirectory.

  (6) Similarity calculation means for calculating similarity between the directory and the created subdirectory, similarity determination means for determining whether the calculated similarity satisfies a predetermined condition, (5) characterized in that it further comprises link creation means for creating a link between the directory and the created subdirectory when the degree of similarity satisfies the predetermined condition as a result of the determination. The information management device described.

  According to the configuration of (6), the information management device creates a link between the directory and the created subdirectory when the degree of similarity between the directory and the created subdirectory satisfies a predetermined condition.

  For this reason, the information management apparatus can further improve the search function in the web directory by creating a link between a directory and a subdirectory having a high degree of similarity.

  (7) In an information management apparatus that manages a URL stored in a directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query, the search result of the search result searched by the search engine Based on the URL and the input query stored in the storage means, the storage means for storing the URL clicked by the user and the input query input by the user terminal among the URLs, A frequency calculating means for calculating the frequency at which a URL stored in a directory is selected, a URL extracting means for extracting a URL that exceeds a predetermined frequency among the calculated frequencies, and a set of the extracted URLs are virtualized. As the directory, similarity calculation means for calculating the similarity with the directory, and the calculated similarity Similarity determination means for determining whether or not a predetermined condition is satisfied, and when the predetermined condition is satisfied as a result of the determination, the extracted URL is stored in the directory where the similarity is calculated. An information management apparatus comprising subdirectory creation means for creating a subdirectory.

  According to the configuration of (7), the information management apparatus extracts URLs exceeding a predetermined frequency, calculates the degree of similarity with the directory using the set of extracted URLs as a virtual directory, and the calculated degree of similarity is When the predetermined condition is satisfied, a subdirectory in which the extracted URL is stored is created in the directory where the similarity is calculated.

  For this reason, the information management device creates a subdirectory in the directory based on the URL and the input query stored in the storage means reflecting the user's social interest. Therefore, the information management apparatus can create a subdirectory that reflects the social interests of the user.

  (8) As a result of the determination, if the predetermined condition is satisfied, the extracted URL is added to the subdirectory, and the added URL overlaps with the URL stored in the directory where the similarity is calculated. The information management apparatus according to (7), further comprising URL merging means for merging URLs to be merged.

  According to the configuration of (8), when the calculated similarity satisfies the predetermined condition, the information management apparatus adds the extracted URL to the subdirectory, and stores the extracted URL and the similarity in the calculated directory. URLs that overlap with the URL to be merged are merged.

  As described above, the information management apparatus adds the extracted URL to the subdirectory based on the URL and the input query stored in the storage unit reflecting the user's social interest and interest, and adds the added URL. Are merged. Therefore, the information management apparatus can add a URL reflecting the social interest of the user to the subdirectory.

  (9) The URL stored in the directory, the search engine that searches the web page, and the user terminal from which the user inputs the search query are managed, and the search query is the URL of the search result searched by the search engine. Among these, in the information management method in which the URL clicked by the user and the input query input by the user terminal are stored in association with each other, the URL is stored in the directory based on the stored URL and the input query. A first link creating step for creating a first directional link to the input query for obtaining the URL as the search result from the URL, and based on the stored URL and the input query, the first The input query is searched by the search engine from the input queries constituting the directed link. Among the search result URLs, a second link creating step for creating a second directional link to the selected URL selected, and the second directional link from the directory in which the URL constituting the first directional link is stored. A transition probability calculating step for calculating a transition probability to the selected URL constituting the directional link, a threshold determining step for determining whether or not the transition probability is larger than a predetermined threshold, and the threshold determining step As a result, when the transition probability is greater than the predetermined threshold, a URL determination step for determining whether or not the selected URL exists in the directory, and a result of the determination by the URL determination step as a result of the selection URL A display order changing step for changing the display order of the selected URL of the directory. Information management method according to claim Rukoto.

  (10) The URL stored in the directory, the search engine that searches the web page, and the user terminal from which the user inputs the search query are managed, and the search query is the URL of the search result searched by the search engine. A computer program executed by a computer of an information management apparatus in which a URL clicked by the user and an input query input by the user terminal are stored in association with each other, the stored URL and the input A first link creating step of creating a first directional link to the input query for obtaining the URL as the search result from a URL stored in the directory based on the query; and the stored URL and Based on the input query, the input queries that constitute the first directed link. A second link creating step for creating a second directional link to the selected URL selected from the URLs of search results obtained by searching the input query from the search engine, and the first directional link. A transition probability calculating step for calculating a transition probability from the directory in which the URL is stored to the selected URL constituting the second directed link, and a threshold for determining whether the transition probability is greater than a predetermined threshold A determination step; and a URL determination step for determining whether or not the selected URL exists in the directory when the transition probability is larger than the predetermined threshold as a result of the determination in the threshold determination step; As a result of the determination in the URL determination step, if the selected URL exists in the directory, the directory is selected. Computer program for executing the display order changing step of changing the display order of the URL to the computer.

  (11) managing a URL stored in a seed directory relating to a specific specialized field, a search engine for searching a web page, and a user terminal where a user inputs a search query, and setting the seed directory to the specific specialized field A specialized information management apparatus, wherein a URL clicked by the user and an input query input by the user terminal are stored in association with each other among URLs of search results obtained by searching the search query by the search engine. Based on the first selection criteria from the URL and the input query by all users stored in the storage means among the URL and the input query stored in the storage means, First extraction means for extracting the URL and the input query related to the seed directory; Based on a second selection criterion, the second extraction means for extracting a URL to be added to the seed directory from the URL extracted by the first extraction means; and the URL extracted by the second extraction means An information management apparatus comprising URL addition means for adding to a seed directory.

  According to the configuration of (11), the information management device extracts the URL and the input query related to the seed directory related to the specific specialized field from the URL and the input query stored in the storage unit based on the first selection criterion. The URL to be added to the directory is extracted from the extracted URL based on the second selection criterion, and the extracted URL is added to the directory.

  For this reason, the information management apparatus can add a URL deeply related to a specific specialized field to the seed directory. Therefore, the information management apparatus can expand the seed directory by suitably specializing the seed directory related to a specific specialized field with the added URL.

  (12) Based on the URL and the input query stored in the storage means, a first directional link to the input query for obtaining the URL as the search result from the URL stored in the seed directory Based on the first link creating means to be created and the URL and the input query stored in the storage means, the input query is searched by the search engine from the input query constituting the first directional link. Among the search result URLs, the second link creating means for creating a second directional link to the selected selected URL, and the second from the seed directory in which the URL constituting the first directional link is stored. A transition probability calculating means for calculating the transition probability to the selected URL constituting the directed link, and the previous calculated by the transition probability calculating means Subdirectory creation means for creating a subdirectory in which the names of input queries constituting the first directional link and the second directional link are assigned to the seed directory based on a transition probability, and the subdirectory creation means (11) The information management apparatus according to (11), further comprising subdirectory addition means for adding the subdirectory created by the step to the seed directory.

  According to the configuration of (12), the information management apparatus is given the names of the input queries that constitute the first directional link and the second directional link based on the transition probability calculated by the transition probability calculating means. Create a subdirectory and add the created subdirectory to the seed directory.

  For this reason, the information management apparatus can add the subdirectory created based on the URL and the input query stored in the storage unit to the seed directory created to search for a specific specialized field. Accordingly, the information management apparatus can expand the seed directory by preferably specializing the seed directory related to a specific field of expertise with the added subdirectory.

  (13) The first link creating unit may include the URL input by the user who has accessed the seed directory and the user who has accessed the seed directory among the URL and the input query stored in the storage unit. Based on the input query input by the terminal, a first directional link to the input query for obtaining the URL as the search result is created from the URL stored in the seed directory, and the second link is created The means includes the URL input by the user accessing the seed directory and the input query input by the user terminal accessing the seed directory among the URL and the input query stored in the storage means. The first directed link based on The information according to (12), wherein a second directional link to a selected URL selected from URLs of search results obtained by searching the input query by the search engine is created from the input query. Management device.

  According to the configuration of (13), the information management apparatus seeds the created subdirectory based on the URL input by the user who has accessed the seed directory and the input query input by the user terminal which has accessed the seed directory. Can be added to the directory.

  (14) The information management apparatus according to (12), further comprising: a selection unit that selects each URL stored in the subdirectory added by the subdirectory addition unit.

  According to the configuration of (14), the information management apparatus selects each URL stored in the added subdirectory. For this reason, the information management apparatus can maintain the URL stored in the subdirectory in a state suitable for a search in a specific specialized field.

  (15) It manages a URL stored in a seed directory related to a specific specialized field, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and the search query is searched by the search engine. Information management apparatus for storing the URL clicked by the user and the input query input by the user terminal in association with each other, and specializing the seed directory in the specific specialized field Information stored in the URL and the input query stored, based on a first selection criterion from the URL stored by all users and the input query by all users. A first extraction step for extracting the URL and the input query related to the seed directory. And a second extraction step for extracting a URL to be added to the seed directory based on a second selection criterion from the URL extracted in the first extraction step, and the second extraction step. And a URL adding step of adding a URL to the seed directory.

  (16) It manages a URL stored in a seed directory related to a specific specialized field, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and the search query is searched by the search engine. Among the URLs of the search results, the URL clicked by the user and the input query input by the user terminal are stored in association with each other, and the information management apparatus computer specializes the seed directory in the specific specialized field The computer program executed by the method according to claim 1, wherein, among the stored URL and the input query, the seed based on a first selection criterion from the stored URL and the input query by all users. Extract the URL and the input query related to the directory A second extraction step of extracting a URL to be added to the seed directory based on a second selection criterion from the URL extracted in the first extraction step, and the second extraction step A computer program for causing a computer to execute a URL adding step of adding the URL extracted by the above to the seed directory.

  (17) An information management device that manages a URL stored in a seed directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and personalizes the seed directory to the user Storage means for storing a URL clicked by the user and an input query input by the user terminal among URLs of search results obtained by searching the search query by the search engine; Of the URL and the input query stored in the storage unit, based on the URL by all users and the input query by all users stored in the storage unit, from the URL stored in the seed directory Create a first directed link to the input query to obtain a URL as the search result The first link creating means, and the URL and the input query stored in the storage means, based on the URL by all users and the input query by all users stored in the storage means, Second link creating means for creating a second directed link to a selected URL selected from among the search results URLs in which the input query is searched by the search engine from the input query constituting one directed link; A transition probability calculating means for calculating a transition probability from the seed directory in which the URL constituting the first directional link is stored to the selected URL constituting the second directional link; and the transition probability calculating means. Based on the calculated transition probability, an input configuring the first directional link and the second directional link in the seed directory A subdirectory creating means for creating a subdirectory to which an Eri name is assigned, and whether or not to add the subdirectory created by the subdirectory creating means to the seed directory based on a subdirectory determination criterion An information management apparatus comprising: a subdirectory determining unit; and a subdirectory adding unit that adds the subdirectory created by the subdirectory creating unit to the seed directory according to a determination result by the subdirectory determining unit. .

  According to the configuration of (17), the information management device creates a subdirectory to which the names of the input queries constituting the first directional link and the second directional link are assigned based on the calculated transition probability. Then, based on the subdirectory determination criterion, it is determined whether or not the created subdirectory is added to the seed directory, and the subdirectory created according to the determination result is added to the seed directory.

  For this reason, the information management apparatus can add a subdirectory personalized for the user operating the user terminal to the seed directory based on the URL and the input query stored in the storage unit. Therefore, the information management apparatus can personalize the directory without requiring an operation from the user.

  (18) The first link creating means includes the URL entered by the user who has accessed the seed directory and the user who has accessed the seed directory among the URL and the input query stored in the storage means. Based on the input query input by the terminal, a first directional link to the input query for obtaining the URL as the search result is created from the URL stored in the seed directory, and the second link is created The means includes the URL input by the user accessing the seed directory and the input query input by the user terminal accessing the seed directory among the URL and the input query stored in the storage means. The first directed link based on The information according to (17), wherein a second directional link to a selected URL selected from URLs of search results obtained by searching the input query by the search engine is created from the input query. Management device.

  According to the configuration of (18), the information management device seeds the created subdirectory based on the URL input by the user who has accessed the seed directory and the input query input by the user terminal which has accessed the seed directory. Can be added to the directory.

  (19) The information management apparatus according to (17), further comprising a sorting unit that sorts each URL added by the subdirectory adding unit and stored in the subdirectory.

  According to the configuration of (19), the information management apparatus selects each URL stored in the added subdirectory. For this reason, the information management apparatus can maintain the URL stored in the subdirectory in a state suitable for a search in a specific specialized field.

  (20) A URL stored in the seed directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and a URL of a search result in which the search query is searched by the search engine An information management method executed by an information management device that stores a URL clicked by the user and an input query input by the user terminal in association with each other and personalizes the seed directory to the user. The URL is searched from the URL stored in the seed directory based on the stored URL and the input query by all users among the stored URL and the input query. Create a first directed link to the input query to obtain as a result The input query that constitutes the first directed link based on the stored URL and the input query by all users among the stored URL and the input query. A second link creating step for creating a second directional link to the selected URL selected from the URLs of search results obtained by searching the input query from the search engine, and the first directional link. Based on the transition probability calculated by the transition probability calculating step for calculating the transition probability from the seed directory where the URL is stored to the selected URL constituting the second directed link, and the transition probability calculating step, Names of input queries constituting the first directional link and the second directional link are attached to the seed directory. A subdirectory creation step for creating the subdirectory, and a subdirectory determination step for determining whether or not to add the subdirectory created in the subdirectory creation step to the seed directory based on a subdirectory determination criterion; An information management method comprising: a subdirectory adding step of adding the subdirectory created by the subdirectory creating step to the seed directory in accordance with a determination result by the subdirectory determining step.

  (21) A URL stored in a seed directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and a URL of a search result in which the search query is searched by the search engine A computer program executed by a computer of an information management device for storing a URL clicked by the user and an input query input by the user terminal in association with each other and personalizing the seed directory to the user The URL is stored in the seed directory based on the stored URL and the input query by all users, and the URL is stored in the seed directory. A second to the input query to get as a search result A first link creating step for creating a directional link; and the first directed link based on the stored URL and the input query by all users among the stored URL and the input query. A second link creating step of creating a second directional link to a selected URL selected from among the URLs of search results in which the input query is searched by the search engine from the input query constituting a directional link; Calculated by a transition probability calculating step for calculating a transition probability from the seed directory in which the URL constituting the first directional link is stored to the selected URL constituting the second directional link, and the transition probability calculating step. Based on the transition probability, the first directional link and the second directional link are configured in the seed directory. A subdirectory creation step for creating a subdirectory to which an input query name is assigned, and whether to add the subdirectory created in the subdirectory creation step to the seed directory based on a subdirectory determination criterion A computer program for causing a computer to execute a subdirectory determining step for determining and a subdirectory adding step for adding the subdirectory created by the subdirectory creating step to the seed directory according to the determination result by the subdirectory determining step.

  According to the present invention, it is possible to provide an information management apparatus, method, and program for creating and displaying a directory reflecting the degree of social interest of a user in a directory-type search method. Further, it is possible to provide an information management apparatus, method, and program for creating a directory that reflects the interest of a specific specialized field or the interest of a specific individual in a directory-type search method.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[Functional configuration of search system 1]
FIG. 1 is a configuration diagram showing a search system 1 according to an embodiment of a search processing apparatus of the present invention. The information management server (information management device) 10, the search server 20, the terminal device 30, and the content server 40 are connected via a communication line 50. The user of the terminal device 30 accesses the search server 20 and acquires a search result by inputting a query (search keyword) for reaching a desired Web page to a predetermined search engine. The user selects a URL listed in the search result and browses a Web page managed by the content server 40.

  The terminal device (user terminal) 30 represents a terminal device such as a personal computer, a mobile phone, a PHS (Personal Handy-phone System: registered trademark), a mobile, a notebook PC, a PDA (Personal Digital Assistant), and the like. Alternatively, any information can be used as long as the information transmitted from the information management server 10 and the search server 20 can be referred to via the communication line 50 that performs wireless communication. In addition, the information management server 10 and the search server 20 may be configured to include a Web server, a DB server, and an application server as necessary, and may be configured as a single server or as separate servers. Also good. The information management server 10 and the search server 20 may be configured as a server group in which one or a plurality of servers are integrated. Furthermore, the information management server 10 manages the terminal device 30 where the user inputs a search query.

[Hardware Configuration of Information Management Server 10]
FIG. 2 is a diagram illustrating an example of a hardware configuration of the information management server 10 according to an embodiment of the present invention. As shown in FIG. 2, a CPU 11 (a plurality of CPUs such as a CPU 12 may be added in a multiprocessor configuration), a communication I / F (I / F: interface) 13, a main memory 14, a display device 15, an I / O A controller 16 and an input device 17 such as a keyboard and a mouse are connected by a bus line BUS to constitute the information management server 10.

  The CPU 11 (CPU 12) is a part that controls the information management server 10 in an integrated manner. The CPU 11 (CPU 12) cooperates with the hardware described above by appropriately reading and executing various programs stored in the hard disk 18a, which will be described later. Various functions related to are realized.

  The communication I / F 13 is a network adapter when the information management server 10 receives various types of input information from the terminal device 30 via the communication line 50 and transmits display contents to the display device 15. . The communication I / F 13 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

  The display device 15 includes a display device such as a cathode ray tube display device (CRT) or a liquid crystal display device (LCD). The I / O controller 16 can be connected to a storage device 18 such as a hard disk 18a and a semiconductor memory 18b. The input device 17 accepts input by an administrator of the information management server 10. The hard disk 18a stores various programs for causing the hardware to function as the information management server 10, a program for executing the functions of the present invention, a table to be described later, and the like. The information management server 10 can also use an external hard disk (not shown) as an external storage device.

  In the above example, the information management server 10 has been mainly described. However, the functions described above can also be realized by installing a program in a computer and operating the computer as a server device. Therefore, the functions realized by the information management server 10 described as an embodiment in the present invention are executed by executing the above-described method by the computer, or by introducing the above-described program into the computer and executing it. Is also feasible.

<First Embodiment>
[Functional configuration of information management server 10]

  FIG. 3 is a functional block diagram of the information management server 10 of the present embodiment. The information management server 10 includes a storage unit 110 mainly configured by the storage device 18 and a control unit 120 mainly configured by the CPU 11 (CPU 12).

  The storage unit 110 includes a click data storage unit 111 and a web directory storage unit 112.

  The click data storage unit (storage unit) 111 selects a query and a URL (Uniform Resource Locator) of a search result searched by a search engine using the query as a search keyword by an arbitrary terminal device 30 operated by the user ( The clicked URL (hereinafter also referred to as “click URL”) is associated with each other and stored as history data. The history data is stored accumulatively as a click data table 111a, for example, as shown in FIG. The click data table 111a illustrated in FIG. 4 stores a query, a click URL, a search rank of the click URL, an IP address of a terminal that has selected the click URL, a selected time, and a session in association with each other. The history data stored in the click data table 111a is information that changes at any time according to the query of the terminal device 30 operated by the user and the selection of the click URL, and reflects the social interest of the user.

  The web directory storage unit 112 stores information on a web directory in the directory type search method and a URL stored in the web directory. FIG. 5 is a schematic diagram showing the web directory W stored in the web directory storage unit 112. The web directory W includes a directory (D1 in FIG. 5), subdirectories (D11 and D12 in FIG. 5) of this directory, and URLs (U11 in FIG. 5) stored in the subdirectory. The names of the directories and subdirectories are given the attributes (categories) of URLs stored in the respective directories and subdirectories. In addition, U11 in FIG. 5 indicates the display order in the descending order from the top when displaying on the display device of the terminal device 30. In the example of FIG. 5, the display order is the order of URL1, URL2, URL3, URL4,. Note that the web directory W, the directory D1, the subdirectories D11 and D12, and the URL U11 shown in FIG. 5 are the same in FIGS.

  The control unit 120 includes a link creation unit 121, a transition probability calculation unit 122, a threshold determination unit 123, a URL determination unit 124, a display order changing unit 125, a URL addition unit 126, a subdirectory creation unit 127, A subdirectory URL adding unit 128, a similarity calculating unit 129, a similarity determining unit 130, and a subdirectory link creating unit 131 are provided.

  Based on the query and the click URL stored in the click data table 111a, the link creation unit (first link creation means, second link creation means) 121 uses this URL from the URL (u) stored in the directory (Dir). A first directional link that is a directional link to a query (q) that is a search keyword for obtaining (u) as a search result is created. Here, the directional link refers to a link having directionality in the link relationship between one object to be linked and the other object.

  In addition, the link creation unit 121 uses the query stored in the click data table 111a and the click URL as a search result URL searched by the search engine using the query as a search keyword from the queries constituting the first directional link. The second directional link to the URL selected by any terminal device 30 operated by the user (hereinafter referred to as the selected URL) is created.

  The transition probability calculation unit (transition probability calculation means, first transition probability calculation means, second transition probability calculation means, third transition probability calculation means) 122 is based on the query and click URL stored in the click data table 111a. The transition probability from the directory storing the URL constituting the first directional link to the selected URL constituting the second directional link is calculated.

  Specifically, the transition probability calculation unit 122 calculates the transition probability (first transition probability) from the URL that constitutes the first directional link created by the link creation unit 121 to the query that constitutes the first directional link. To do. Specifically, when the click URL (u) in which the first directional link is created is selected, the transition probability (first transition probability) P when the click URL (u) is based on the selection query (q). (Q | u) is calculated.

  Further, the transition probability calculation unit 122 generates the second directional from the selected query (q) that configures the first directional link and the second directional link based on the query and the click URL stored in the click data table 111a. The transition probability (second transition probability) to the selected URL (u ′) constituting the link is calculated. Specifically, when the query (q) constituting the first directed link is selected, the search result searched by the selection query (q) is selected by an arbitrary terminal device 30 operated by the user. The transition probability (second transition probability) P (u ′ | q) is calculated when the selected URL is the selected URL (u ′).

  The transition probability calculation unit 122 also stores the URL (u) stored in the directory (Dir) based on the calculated first transition probability P (q | u) and second transition probability P (u ′ | q). ) To the selected URL (u ′). Specifically, the transition probability P (u ′ | Dir), that is, the sum of P (q | u) * P (u ′ | q) for each selected URL (u ′) is calculated.

  The transition probability calculation unit 122 configures the first directional link and the second directional link from the directory (Dir) in which the URL (u) is stored based on the first transition probability P (q | u). A transition probability (third transition probability) to the query (q) is calculated. Specifically, the third transition probability P (q | Dir), that is, the first transition probability P (q | u) for each query (q) constituting the first directional link and the second directional link. Calculate the sum of.

  Further, the transition probability calculation unit 122 sorts the calculated first transition probability, second transition probability, and third transition probability. Specifically, the calculated first transition probability, second transition probability, and third transition probability are sorted in descending order or ascending order.

  The threshold determination unit (first threshold determination unit, second threshold determination unit, third threshold determination unit) 123 has a transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 122 greater than the first threshold. Determine whether it is larger.

Further, the threshold determination unit 123 determines whether or not the third transition probability P (q | Dir) calculated by the transition probability calculation unit 122 is larger than the second threshold.
Further, the threshold determination unit 123 determines whether or not the transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 122 is larger than the third threshold.
Here, the first threshold value, the second threshold value, and the third threshold value can be set to arbitrary values.

  As a result of the determination by the threshold determination unit 123, if the transition probability P (u ′ | Dir) is larger than the first threshold, the URL determination unit 124 stores the selected URL (u ′) in the directory (Dir). Determine if it exists.

  When the selected URL (u ′) exists in the directory (Dir) as a result of the determination by the URL determining unit 124, the display order changing unit 125 displays the selected URL (u ′) in the directory (Dir). Change the ranking.

  If the selected URL (u ′) does not exist in the directory (Dir) as a result of the determination by the URL determining unit 124, the URL adding unit 126 adds the selected URL (u ′) to the directory (Dir). To do.

  When the third transition probability P (q | Dir) is larger than the second threshold as a result of the determination by the threshold determination unit 123, the subdirectory creating unit 127 uses the first directory as the subdirectory of the directory (Dir). A subdirectory (SubDir) to which the name of the query (q) constituting the directional link and the second directional link is assigned is created. The subdirectory thus created is also called a click directory.

  If the transition probability P (u ′ | Dir) is larger than the third threshold as a result of the determination by the threshold determination unit 123, the subdirectory URL adding unit 128 generates the subdirectory created by the subdirectory creation unit 127. The selection URL (u ′) is added in (SubDir).

  The similarity calculation unit 129 calculates the similarity between the directory (Dir) in which the URL (u) is stored and the subdirectory (SubDir) created by the subdirectory creation unit 127. Specifically, each URL (u) stored in the directory (Dir) and each selected URL (u ′) stored in the subdirectory (SubDir) are sequentially matched to obtain the similarity sim (Dir , SubDir). Here, Cosine similarity, Dice similarity, Jaccard similarity, etc. can be applied as a measure of similarity sim.

  The similarity determination unit 130 determines whether or not the similarity sim (Dir, SubDir) calculated by the similarity calculation unit 129 satisfies a predetermined condition. Specifically, as a predetermined condition, for example, a directory having the highest similarity sim (Dir, SubDir) among all directories, a similarity sim (Dir, SubDir) is greater than a predetermined threshold, or all directories It is possible to apply conditions such as the directory having the highest similarity sim (Dir, SubDir) and the similarity sim (Dir, SubDir) being larger than a predetermined threshold.

  When the similarity sim (Dir, SubDir) satisfies a predetermined condition as a result of the determination by the similarity determination unit 130, the subdirectory link creation unit 131 is created by the directory (Dir) and the subdirectory creation unit 127. Create a link with (SubDir).

[Process flow]
Next, the processing flow of the information management server 10 of this embodiment will be described with reference to the flowcharts of FIGS. 9 and 10 are diagrams showing the directed link created by the link creating unit 121. FIG. FIGS. 11 to 13 are schematic diagrams illustrating processing executed by the information management server 10 of this embodiment for the web directory W.

  Step S1: The link creation unit 121 obtains the URL (u) as a search result from the URL (u) stored in the directory (Dir) based on the query and the click URL stored in the click data table 111a. A first directional link that is a directional link to the query (q) that is a search keyword is created. Specifically, as shown in FIG. 9, the URL (u) stored in the directory (URL1, URL2, URL3, and URL4 in FIG. 9) to the query (q) (Q1, Q2, Q3 in FIG. 9 and Create a first directed link to Q4). Here, the arrow of the directed link indicates the direction of the link. In the example of FIG. 9, the first directed link from URL1 to queries Q1 and Q2 is created. Also, a first directed link from URL2 to query Q2 is created. A first directed link from URL3 to queries Q2, Q3, and Q4 is created. In addition, a first directed link from URL4 to query Q4 is created.

  Step S2: The link creation unit 121 searches for this query (q) from the query (q) constituting the first directional link created in Step S1, based on the URL and query stored in the click data table 111a. A second directional link to a selected URL (u ′) selected by an arbitrary terminal device 30 operated by a user among URLs of search results searched by a search engine as a keyword is created. Specifically, as shown in FIG. 10, the selected URL (u ′) (in FIG. 10) is selected from the query (q) (Q1, Q2, Q3, and Q4 in FIG. 10) in which the first directed link is created. URL 11, URL 12, URL 13, and URL 14) are created. In the example of FIG. 10, a second directed link from the query Q1 to the URL 11 is created. In addition, a second directional link from the query Q2 to the URL 11, URL 12, and URL 13 is created. In addition, a second directional link from the query Q3 to the URL 13 is created. In addition, a second directional link from the query Q4 to the URL 14 is created. As described above, the bipartite graph composed of the URL (u), the query (q), the selected URL (u ′), the first directional link, and the second directional link is created by the processing of step S1 and step S2. Is done.

  Step S3: The transition probability calculation unit 122 calculates a transition probability P (u ′ | Dir). Specifically, when the URL (u) constituting the first directional link is selected in the first directional link created in step S1, the URL (u) is based on the selected query (q). The transition probability (first transition probability) P (q | u) is calculated. Further, when the query (q) constituting the first directional link and the second directional link is selected in the second directional link created in step S2, the transition probability calculation unit 122 selects this query (q ) To calculate the transition probability (second transition probability) P (u ′ | q) when the arbitrarily selected URL is the selected URL (u ′). Then, a transition probability P (u ′ | Dir) is calculated based on the first transition probability P (q | u) and the second transition probability P (u ′ | q). That is, the sum of P (q | u) * P (u ′ | q) is calculated for each selected URL (u ′).

  Step S4: The transition probability calculation unit 122 sorts the transition probabilities P (u ′ | Dir) for each selected URL (u ′) calculated in Step S3 in descending order. In this embodiment, the transition probabilities P (u ′ | Dir) are sorted in descending order. However, the present invention is not limited to this, and may be sorted in ascending order or may be sorted by other methods.

  Step S5: The threshold determination unit 123 determines whether or not the transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 122 is larger than the first threshold. When the transition probability P (u ′ | Dir) is larger than the first threshold (Yes), the process proceeds to step S6. On the other hand, when the transition probability P (u ′ | Dir) is equal to or less than the first threshold value (No), the series of processing ends.

  Step S6: The URL determination unit 124 determines whether or not the selected URL (u ′) exists in the directory (Dir). If the selected URL (u ′) exists in the directory (Dir) (Yes), the process proceeds to step S7. On the other hand, if the selected URL (u ′) does not exist in the directory (Dir) (No), the process proceeds to step S8.

  Step S7: When the selected URL (u ′) exists in the directory (Dir), the display order changing unit 125 changes the display order of the selected URL (u ′) in the directory (Dir). Specifically, the display order of U11 in the directory D11 is changed as shown in FIG. 11A, the display order URL1, URL2, URL3, URL4... Shown in FIG. 5 is changed to the display order URL4, URL1, URL2, URL3.

  As described above, the information management server 10 changes the display order of the selected URL (u ′) based on the query and the click URL stored in the click data table 111a reflecting the user's social interest. To do. Therefore, the information management server 10 can reflect the user's social interest in the display order of the selected URL (u ′).

  Step S8: If the selected URL (u ′) does not exist in the directory (Dir), the URL adding unit 126 adds the selected URL (u ′) to the directory (Dir). Specifically, as shown in FIG. 11B, in addition to the existing URL1, URL2, URL3, and URL4, URL5 is added as a selected URL (u ′) to U11 in the directory D11. As described above, the information management server 10 stores the selected URL (u ′) in the directory (Dir) based on the query and the click URL stored in the click data table 111a reflecting the user's social interest. Add in. Therefore, the information management server 10 can add the selection URL (u ′) reflecting the social interest of the user to the directory (Dir).

  Step S9: The transition probability calculation unit 122 calculates a fourth transition probability P (q | Dir). Specifically, in the first directed link created in step S2, the first query for each selection query (q) is based on the first transition probability P (q | u) calculated in step S3. The sum total of the transition probabilities P (q | u) is calculated. Thereby, the fourth transition probability P (q | Dir) can be calculated.

  Step S10: The transition probability calculation unit 122 sorts the fourth transition probabilities P (q | Dir) for each selection query (q) calculated in step S9 in descending order. In the present embodiment, the fourth transition probabilities P (q | Dir) are sorted in descending order. However, the fourth transition probabilities P (q | Dir) may be sorted in ascending order or may be sorted by other methods.

  Step S11: The threshold determination unit 123 determines whether or not the fourth transition probability P (q | Dir) calculated by the transition probability calculation unit 122 is greater than the second threshold. When the fourth transition probability P (q | Dir) is larger than the second threshold (Yes), the process proceeds to step S12. On the other hand, when the fourth transition probability P (q | Dir) is equal to or smaller than the second threshold (No), the process of this flowchart is terminated.

  Step S12: When the fourth transition probability P (q | Dir) is larger than the second threshold, the subdirectory creating unit 127 uses the first directional link and the second directional as a subdirectory of the directory (Dir). A subdirectory (SubDir) to which the name of the query (q) constituting the link is assigned is created. Specifically, as shown in FIG. 12, a subdirectory D21 to which the name of the query Q2 is assigned is created as a subdirectory of the subdirectory D11.

  As described above, the information management server 10 includes the first directional link and the second directional link based on the query and the click URL stored in the click data table 111a reflecting the user's social interest. Is created as a subdirectory of the directory (Dir). Therefore, the information management server 10 can create a subdirectory (SubDir) to which the name of the query (q) reflecting the social interest of the user is assigned as a subdirectory of the directory (Dir).

  Step S13: The threshold determination unit 123 determines whether or not the transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 122 is greater than the third threshold. When the transition probability P (u ′ | Dir) is larger than the third threshold (Yes), the process proceeds to step S14. On the other hand, when the transition probability P (u ′ | Dir) is equal to or less than the third threshold value (No), the series of processing ends.

  Step S14: If the transition probability P (u ′ | Dir) is larger than the third threshold in step S13, the subdirectory URL adding unit 128 creates the subdirectory (SubDir) created by the subdirectory creating unit 127. The selection URL (u ′) is added to the inside. Specifically, as shown in FIG. 12, URL11, URL12, URL13, URL14... Are added as selection URLs (u ′) to U21 in the subdirectory D21.

  As described above, the information management server 10 converts the selected URL (u ′) to the subdirectory (SubDir) based on the query and the click URL stored in the click data table 111a reflecting the user's social interest. ). Therefore, the information management server 10 can add the selection URL (u ′) reflecting the social interest / interest of the user in the subdirectory (SubDir).

  After finishing the process of step S14, the information management server 10 proceeds to the processes of step S15 and step S18.

  Step S15: The similarity calculation unit 129 calculates the similarity sim (Dir, SubDir) between the directory (Dir) in which the URL (u) is stored and the subdirectory (SubDir) created by the subdirectory creation unit 127. .

  Step S16: The similarity determination unit 130 determines whether the similarity sim (Dir, SubDir) calculated in Step S15 satisfies a predetermined condition. When the similarity sim (Dir, SubDir) satisfies a predetermined condition (Yes), the process proceeds to step S17. On the other hand, when the similarity sim (Dir, SubDir) does not satisfy the predetermined condition (No), the series of processing ends.

  Step S17: When the similarity sim (Dir, SubDir) satisfies the predetermined condition in step S16, the subdirectory link creation unit 131 determines the directory (Dir) and the subdirectory (SubDir) created in step S12. Create a link for Specifically, as shown in FIG. 13, a link L between the subdirectory D12 and the subdirectory D21 is created.

  Thus, the information management server 10 can further improve the search function in the web directory by creating a link between the directory and the subdirectory having a high similarity.

  Step S18: Based on the query and click URL stored in the click data table 111a, the link creating unit 121 performs a search for obtaining the URL (u) as a search result from the URL (u ′) stored in the subdirectory. A first directional link that is a directional link to a query (q ′) that is a keyword is created.

  Step S19: The link creation unit 121 obtains this query (q) from the query (q) constituting the first directional link created in Step S18 based on the query and the click URL stored in the click data table 111a. A second directional link to a selected URL (u ′) selected by an arbitrary terminal device 30 operated by the user among URLs of search results searched by the search engine as a search keyword is created.

  Step S20: The transition probability calculation unit 122 calculates a transition probability P (u ″ | Dir). Specifically, when the URL (u ′) is selected by the arbitrary terminal device 30 operated by the user in the first directional link created in step S18, the URL (u ′) is selected by the selection query (q The first transition probability P (q ′ | u ′) which is') is calculated. Further, the transition probability calculation unit 122, when the selection query (q ′) is selected in the second directed link created in step S19, among the search results searched by the selection query (q ′), A second transition probability P (u ″ | q ′) in which the selected URL is the selected URL (u ″) is calculated. Based on the first transition probability P (q ′ | u ′) and the second transition probability P (u ″ | q ′), the transition probability P (u ″ | Dir), that is, each selected URL ( The sum of P (q ′ | u ′) * P (u ″ | q ′) for u ″) is calculated.

  Step S21: The transition probability calculation unit 122 sorts the transition probabilities P (u ″ | Dir) for each selected URL (u ″) calculated in step S20 in descending order. In this embodiment, the transition probabilities P (u ″ | Dir) are sorted in descending order. However, the present invention is not limited to this, and may be sorted in ascending order or may be sorted by other methods.

  Step S22: The threshold determination unit 123 determines whether or not the transition probability P (u ″ | Dir) calculated by the transition probability calculation unit 122 is greater than the first threshold. When the transition probability P (u ″ | Dir) is larger than the first threshold (Yes), the process proceeds to step S23. On the other hand, when the transition probability P (u ″ | Dir) is equal to or less than the first threshold (No), the processing of this flowchart is terminated.

  Step S23: The URL determination unit 124 determines whether or not the selected URL (u ″) exists in the directory (Dir). If the selected URL (u ″) exists in the directory (Dir) (Yes), the process proceeds to step S24. On the other hand, if the selected URL (u ″) does not exist in the directory (Dir) (No), the process proceeds to step S24.

  Step S24: When the selected URL (u ″) exists in the directory (Dir), the display order changing unit 125 changes the display order of the selected URL (u ″) in the directory (Dir). Thus, by changing the display order of the selected URLs in the subdirectory, the same effect as the process in step S7 can be achieved for the selected URLs in the subdirectory.

  Step S25: If the selected URL (u ″) does not exist in the directory (Dir), the URL adding unit 126 adds the selected URL (u ″) to the directory (Dir). Thus, by adding the selection URL to the subdirectory, the same effect as the process in step S8 can be achieved for the subdirectory.

  As described above, the information management server 10 of this embodiment can provide a directory-type search method that reflects the social interests of users. Therefore, the search function of the search service by the directory type search method is improved, and the use of the search service by the directory type search method by the user can be increased.

<Second Embodiment>
Next, a second embodiment of the present invention will be described. About 2nd Embodiment, a different point from 1st Embodiment is mainly demonstrated, the same code | symbol is attached | subjected about the structure similar to 1st Embodiment, and description is abbreviate | omitted.

[Functional configuration of information management server 10]
FIG. 14 is a functional block diagram of the information management server 10 of the present embodiment. The control unit 120 of the information management server 10 includes a frequency calculating unit 141, a URL extracting unit 142, a similarity calculating unit 143, a similarity determining unit 144, a subdirectory creating unit 145, and a URL merging unit 146. Prepare. Since other functional configurations are the same as those in the first embodiment, description thereof is omitted.

  The frequency calculation unit 141 calculates the frequency at which the URL (u) stored in the directory (Dir) is selected based on the query and the click URL stored in the click data table 111a.

  The URL extraction unit 142 extracts URLs that exceed a predetermined frequency among the frequencies calculated by the frequency calculation unit 141.

  The similarity calculation unit 143 calculates the similarity with the directory (Dir) using the set of URLs extracted by the URL extraction unit 142 as a virtual directory (C). Specifically, each URL stored in the virtual directory and each URL (u) stored in the directory (Dir) are sequentially matched to calculate the similarity sim (C, Dir). Here, Cosine similarity, Dice similarity, Jaccard similarity, etc. can be applied as a measure of similarity sim.

  The similarity determination unit 144 determines whether or not the similarity sim (C, Dir) calculated by the similarity calculation unit 143 satisfies a predetermined condition. Specifically, the directory having the highest similarity sim (C, Dir) among all directories, the similarity sim (C, Dir) is greater than a predetermined threshold, or the similarity sim ( It is possible to apply such a condition that the directory is the largest directory (C, Dir) and the similarity sim (C, Dir) is larger than a predetermined threshold.

  When the predetermined condition is satisfied as a result of the determination by the similarity determination unit 144, the subdirectory creation unit 145 is extracted by the URL extraction unit 142 as a subdirectory of the directory for which the similarity sim (C, Dir) has been calculated. A subdirectory (SubDir) in which the URL is stored is created.

  When the predetermined condition is satisfied as a result of the determination by the similarity determination unit 144, the URL merging unit 146 is extracted by the URL stored in the directory in which the similarity sim (C, Dir) is calculated and the URL extraction unit 142 Merge URLs that overlap with URLs. Note that, as a result of determination by the similarity determination unit 144, processing executed when a predetermined condition is satisfied is not limited to the above-described subdirectory creation unit 145 and URL merging unit 146, but a virtual directory (C) and directory Other processes that can be associated to display (Dir) may be applied.

Next, the flow of processing of the information management server 10 of this embodiment will be described with reference to the flowchart shown in FIG.
Step S31: The frequency calculation unit 141 calculates the frequency at which the URL (u) stored in the directory (Dir) is selected based on the query and the click URL stored in the click data table 111a.

Step S32: The URL extraction unit 142 extracts URLs exceeding a predetermined frequency from the frequencies calculated in Step S31.
Step S33: The similarity calculation unit 143 calculates the similarity with the directory (Dir) using the set of URLs extracted in step S32 as a virtual directory (C).

  Step S34: The similarity determination unit 144 determines whether or not the similarity sim (C, Dir) calculated in step S33 satisfies a predetermined condition. If the similarity sim (C, Dir) satisfies a predetermined condition (Yes), the process proceeds to step S35 and step S36. On the other hand, when the similarity sim (C, Dir) does not satisfy the predetermined condition (No), the series of processing ends.

  Step S35: The subdirectory creation unit 145 is extracted in step S32 as a subdirectory of the directory for which the similarity sim (C, Dir) is calculated because the result of the determination in step S34 is that the predetermined condition is satisfied. A subdirectory (SubDir) in which the URL is stored is created. Then, the process of this flowchart is complete | finished.

  As described above, the information management server 10 changes the subdirectory (SubDir) to the directory (Dir) based on the query and the click URL stored in the click data table 111a reflecting the social interest of the user. To create. Therefore, the information management server 10 can create a subdirectory (SubDir) reflecting the social interest of the user.

  Step S36: The URL merging unit 146 adds the URL extracted in Step S32 to the subdirectory (SubDir) because the result determined in Step S34 is that the predetermined condition is satisfied, and the similarity with the added URL The URLs that overlap with the URLs stored in the directory where sim (C, Dir) is calculated are merged. Then, the process of this flowchart is complete | finished.

  As described above, the information management server 10 stores the extracted URL in the subdirectory (SubDir) based on the query and the click URL stored in the click data table 111a reflecting the user's social interest. Add and merge the added URLs. Therefore, the information management server 10 can be added in a subdirectory (SubDir) reflecting the social interest of the user.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

<Third Embodiment>
Next, a third embodiment of the present invention will be described.

[Functional configuration of information management server 10]
FIG. 16 is a functional block diagram of the information management server 10 of this embodiment. The storage unit 110 of the information management server 10 includes a click data storage unit 111, a web directory storage unit 112, an access log storage unit 113, a selection criterion storage unit 114, and a sorting selection criterion storage unit 115. The click data storage unit 111 is the same as that in the first and second embodiments, and a description thereof will be omitted.

  The web directory storage unit 112 stores information on a web directory in the directory type search method and a URL stored in the web directory. FIG. 17 is a schematic diagram showing the web directory W2 stored in the web directory storage unit 112. The web directory W2 includes a directory (D3 in FIG. 17), subdirectories (D31 and D32 in FIG. 17) of this directory, and URLs (U31 in FIG. 17) stored in the subdirectory. The names of the directories and subdirectories are given the attributes (categories) of URLs stored in the respective directories and subdirectories. Moreover, U31 in FIG. 17 indicates the display order when displaying on the display device of the terminal device 30 in descending order from the top. In the example of FIG. 17, the display order is the order of URL 31, URL 32, URL 33, URL. Note that the web directory W2, the directory D31, the subdirectories D31 and D32, and the URL U31 illustrated in FIG. 17 are the same in FIGS.

  Here, in the third embodiment, a directory to be specialized (for example, corresponding to a subdirectory D31 described later) is particularly referred to as a seed directory. The seed directory is created and stored in the web directory storage unit 112 in advance.

  In the present embodiment, the information management server 10 mainly aims to specialize a seed directory related to a specific specialized field.

  The access log storage unit 113 stores an access log when the terminal device 30 accesses the web directory W2, that is, each of the directory D31 and the subdirectories D31 and D32. The access log is stored in association with, for example, a session, URL, IP address and the like.

  The selection criterion storage unit 114 stores a first selection criterion used by the extraction unit 151 to extract a click URL and a query related to the seed directory from the click data storage unit 111. In addition, the selection criterion storage unit 114 stores a second selection criterion used for extracting a click URL to be added to the seed directory from the URL extracted by the extraction unit 151 based on the first selection criterion.

Here, the first selection criterion and the second selection criterion will be described in detail.
For example, the following conditions 1 to 3 can be used as the first selection criterion in order to extract the click URL and the query related to the seed directory from the click data storage unit 111.

(Condition 1) The click URL and the query stored in the click data storage unit 111 are input by the terminal device 30 when the seed directory is accessed by the terminal device 30, and the query is searched by the search engine. Among the search result URLs, the click URL clicked by the user operating the terminal device 30 and the query input by the terminal device 30.
(Condition 2) The click URL and query are stored in the click data storage unit 111 within the specified period for the specified number of times.
(Condition 3) The access to the seed directory by the terminal device 30 is more than the specified number of times.

  For example, the following conditions 4 to 7 can be used as the second selection criterion in order to extract the click URL to be added to the seed directory from the click URL extracted by the extraction unit 151 based on the first selection criterion.

(Condition 4) Three URLs satisfying the condition 1 and having a high frequency of selection of click URLs within a specified period (for example, 1 hour) and the frequency is equal to or more than the specified number.
(Condition 5) Condition 2 is satisfied, and Condition 4 or 6 is satisfied.
(Condition 6) The terminal device 30 accesses the seed directory a specified number of times (first specified number) or more, and selects a click URL stored in the seed directory for a specified number of times (second specified number).
(Condition 7) Conditions 4 to 6 are mixed at a predetermined mixing ratio. For example, the mixing ratio can be set so that the condition 4 is 0.2, the condition 5 is 0.5, and the condition 6 is 0.3, assuming that the entire mixing ratio of the conditions 4 to 6 is 1.

  The selection criterion storage unit 115 deletes URLs of sites with undesirable contents as subdirectories added to the seed directory out of click URLs (links) stored in subdirectories added to the seed directory. A list of selection criteria is stored. The URL of an undesired content site is, for example, a URL of a site that is against public order and morals.

  The control unit 120 includes an extraction unit 151, a URL addition unit 152, a link creation unit 153, a transition probability calculation unit 154, a threshold determination unit 155, a subdirectory addition unit 156, and a sorting selection unit 157. In the following description of this embodiment, as a specific example of the seed directory, the subdirectory D31 is a seed directory to be expanded, and the subdirectory D31 is appropriately referred to as the seed directory D31.

  The extraction unit 151 extracts a URL and an input query related to the seed directory D31 in order to specialize the seed directory D31 based on the first selection criterion from the query and the click URL stored in the click data storage unit 111. To do. Then, a URL to be added to the seed directory D31 is extracted from the extracted URL based on the second selection criterion. The extraction unit 151 preferably uses a query by all users and a click URL by all users among the URLs and input queries stored in the click data storage unit 111.

  The URL adding unit 152 adds the click URL extracted by the extracting unit 151 based on the second selection criterion to the seed directory D31.

  The link creation unit 153 is a search keyword for obtaining the URL (u) as a search result from the URL (u) stored in the directory (Dir) based on the query and the click URL stored in the click data table 111a. A first directional link that is a directional link to a query (q) is created. At this time, as described above, in the present embodiment, a case where the seed directory D31 is used as a specific example of the directory (Dir) will be described. Here, the link creation unit 153 is input by the terminal device 30 having accessed the click URL and the seed directory D31 input by the user who has accessed the seed directory D31 among the queries and click URLs stored in the click data table 111a. It is preferable to use a query.

  Based on the query and the click URL stored in the click data table 111a, the transition probability calculation unit 154 changes from the directory in which the URL that configures the first directional link is stored to the selected URL that configures the second directional link. The transition probability is calculated. Here, the link creation unit 153 is input by the terminal device 30 having accessed the click URL and the seed directory D31 input by the user who has accessed the seed directory D31 among the queries and click URLs stored in the click data table 111a. It is preferable to use a query.

  Specifically, the transition probability calculation unit 122 calculates the transition probability (first transition probability) from the URL that constitutes the first directional link created by the link creation unit 121 to the query that constitutes the first directional link. To do. Specifically, when the click URL (u) in which the first directional link is created is selected, the transition probability (first transition probability) P when the click URL (u) is based on the selection query (q). (Q | u) is calculated.

  In addition, the transition probability calculation unit 154 generates the second directional from the selected query (q) constituting the first directional link and the second directional link based on the query and the click URL stored in the click data table 111a. The transition probability (second transition probability) to the selected URL (u ′) constituting the link is calculated. Specifically, when the query (q) constituting the first directed link is selected, the search result searched by the selection query (q) is selected by an arbitrary terminal device 30 operated by the user. The transition probability (second transition probability) P (u ′ | q) is calculated when the selected URL is the selected URL (u ′).

  In addition, the transition probability calculation unit 154 generates a directory (Dir) in which URL (u) is stored based on the calculated first transition probability P (q | u) and second transition probability P (u ′ | q). ) To the selected URL (u ′). Specifically, the transition probability P (u ′ | Dir), that is, the sum of P (q | u) * P (u ′ | q) for each selected URL (u ′) is calculated.

  Further, the transition probability calculation unit 154 configures the first directional link and the second directional link from the directory (Dir) in which the URL (u) is stored based on the first transition probability P (q | u). A transition probability (third transition probability) to the query (q) is calculated. Specifically, the third transition probability P (q | Dir), that is, the first transition probability P (q | u) for each query (q) constituting the first directional link and the second directional link. Calculate the sum of.

The threshold determination unit 155 determines whether or not the transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 154 is greater than the first threshold.
Further, the threshold determination unit 123 determines whether or not the third transition probability P (q | Dir) calculated by the transition probability calculation unit 154 is larger than the second threshold.

  When the third transition probability P (q | Dir) is larger than the second threshold as a result of the determination by the threshold determination unit 155, the subdirectory adding unit 156 determines the first directional link and the second directional link. Is created and added as a subdirectory of the seed directory. The subdirectory thus created is also called a click directory. That is, the subdirectory adding unit 156 adds the click directory as a subdirectory to the seed directory.

  As described above, according to the information management server 10 of this embodiment, the URL related to the seed directory can be extracted from the click data storage unit 111 and added to the seed directory. A click directory can be added to the seed directory based on the click data storage unit 111.

  Therefore, the information management server 10 can add URLs and subdirectories that are deeply related to a specific specialized field to the seed directory. Therefore, the information management server 10 can expand the seed directory by suitably specializing the seed directory related to a specific specialized field with the added URL and subdirectory.

  The selection selecting unit 157 selects each URL (link) stored in the click directory added to the seed directory D31, based on the access log stored in the access log storage unit 113. Specifically, for example, a URL that has been accessed a predetermined number or less is deleted from the click directory.

  Further, the sorting selection unit 157 performs selection of each URL (link) stored in the click directory added to the seed directory D31 based on the sorting criteria stored in the sorting selection storage unit 115. Specifically, for example, a site having an undesired content among URLs (links) stored in a subdirectory added to the seed directory D31 based on a list of selection criteria stored in the selection criteria storage unit 115, for example. Is deleted.

  As described above, the selection selecting unit 157 performs selection of each URL (link) stored in the click directory added to the seed directory D31. For this reason, the information management server 10 can maintain the URL stored in the click directory in a state suitable for a search in a specific specialized field.

[Process flow]
Next, the processing flow of the information management server 10 of this embodiment will be described with reference to the flowcharts of FIGS. 18 and 19. 20 to 22 are schematic diagrams illustrating processing executed by the information management server 10 according to the present embodiment for the web directory W2.

FIG. 18 is a flowchart showing the flow of processing of the information management server 10 of this embodiment.
Step S41: The extraction unit 151 extracts the URL and the input query related to the seed directory D31 from the URL and the input query stored in the click data storage unit 111 based on the first selection criterion.

  Step S42: The extraction unit 151 extracts a URL to be added to the seed directory D31 from the URL extracted in step S41 based on the second selection criterion.

  Step S43: The URL adding unit 152 adds the URL extracted based on the second selection criterion by the extracting unit 151 in Step S42 to the seed directory D31. Specifically, as shown in FIG. 21, URL 34, URL 35, and URL 36 are added in the seed directory D31 (see FIG. 20) before the URL is added.

  Step S44: The link creation unit 153, the transition probability calculation unit 154, and the threshold determination unit 155 execute click directory creation processing. The click directory creation process will be described in detail with reference to FIG.

  Step S45: The subdirectory adding unit 156 creates a subdirectory (SubDir) to which the name of the query (q) constituting the first directional link and the second directional link is assigned based on the result of the click directory creation process. And add it as a subdirectory of the seed directory. That is, the subdirectory adding unit 156 adds the click directory as a subdirectory to the seed directory. Specifically, as shown in FIG. 22, the click directory D41 is added as a subdirectory of the seed directory D31. The click directory D41 stores URL41, URL42, and URL43.

  Step S46: The selection selection unit 157 is stored in the click directory added to the seed directory D31 based on the access log stored in the access log storage unit 113 or the selection criterion stored in the selection criterion storage unit 115. Each URL (link) is selected, and a series of processing ends.

FIG. 19 is a flowchart showing the flow of click directory creation processing.
Step S51: The link creation unit 153 obtains the URL (u) as a search result from the URL (u) stored in the directory (Dir) based on the query and the click URL stored in the click data table 111a. A first directional link that is a directional link to the query (q) that is a search keyword is created.

  Step S52: The link creation unit 153 retrieves this query (q) from the query (q) constituting the first directional link created in step S51 based on the URL and query stored in the click data table 111a. A second directional link to a selected URL (u ′) selected by an arbitrary terminal device 30 operated by a user among URLs of search results searched by a search engine as a keyword is created. As described above, the bipartite graph composed of the URL (u), the query (q), the selected URL (u ′), the first directional link, and the second directional link is created by the processing of step S51 and step S52. Is done.

  Step S53: The transition probability calculation unit 154 calculates a transition probability P (u ′ | Dir). Specifically, when the URL (u) constituting the first directional link is selected in the first directional link created in step S51, the URL (u) is based on the selection query (q). The transition probability (first transition probability) P (q | u) is calculated. Further, when the query (q) constituting the first directional link and the second directional link is selected in the second directional link created in step S52, the transition probability calculation unit 154 selects this query (q ) To calculate the transition probability (second transition probability) P (u ′ | q) when the arbitrarily selected URL is the selected URL (u ′). Then, a transition probability P (u ′ | Dir) is calculated based on the first transition probability P (q | u) and the second transition probability P (u ′ | q). That is, the sum of P (q | u) * P (u ′ | q) is calculated for each selected URL (u ′).

  Step S54: It is determined whether or not the transition probability P (u ′ | Dir) calculated by the transition probability calculation unit 154 is greater than the first threshold value. When the transition probability P (u ′ | Dir) is larger than the first threshold (Yes), the process proceeds to step S55. On the other hand, when the transition probability P (u ′ | Dir) is equal to or less than the first threshold value (No), the series of processing ends.

  Step S55: The transition probability calculation unit 154 calculates a fourth transition probability P (q | Dir). Specifically, in the first directional link created in step S51, the first query for each selection query (q) is based on the first transition probability P (q | u) calculated in step S53. The sum total of the transition probabilities P (q | u) is calculated. Thereby, the fourth transition probability P (q | Dir) can be calculated.

  Step S56: The threshold determination unit 155 determines whether or not the fourth transition probability P (q | Dir) calculated by the transition probability calculation unit 154 is greater than the second threshold. If the fourth transition probability P (q | Dir) is larger than the second threshold (Yes), the process proceeds to step S57. On the other hand, when the fourth transition probability P (q | Dir) is equal to or smaller than the second threshold (No), the process of this flowchart is terminated.

  Step S57: When the fourth transition probability P (q | Dir) is larger than the second threshold, the subdirectory adding unit 156 sets the first directional link and the second directional as a subdirectory of the directory (Dir). A subdirectory (SubDir) to which the name of the query (q) constituting the link is assigned is created, and a series of processing ends.

[Specific example of web directory W2]
Next, a specific example of the web directory W2 according to the present embodiment will be described with reference to FIGS.
23 to 26 are diagrams illustrating specific display examples of the web directory W2. As shown in FIG. 23, the display screen 60 of the web directory W2 includes each seed directory 61 constituting the web directory W2, and a registration site 62 that is a URL stored (linked) in the seed directory 61. A search word input field 63 for inputting a query and a search button 64 are included.

  An added URL 65 shown in FIG. 24 is a URL added to the seed directory 61 by the URL adding unit 152 (corresponding to step S43 in FIG. 18).

  FIG. 25 is a diagram showing a specific example of the click directory (subdirectory) 66 created and added to the seed directory 61a in the seed directory 61 shown in FIG. As shown in FIG. 25, since the category name of the seed directory 61a is “Diet”, the created click directory 66 is given a category name highly relevant to “Diet”. In FIG. 25, a click directory 66 to which category names of “calorie control”, “diet recipe”, “diet experience”, and “fitness” are assigned is created.

  FIG. 26 is a diagram showing URLs stored in the click directory 66a of the category name “calorie control” in the click directory 66 shown in FIG. As shown in FIG. 26, the URL 68 stored in the “calorie control” category name click directory 66a is given the names “calorie control dictionary”, “calculation of calories on the Internet”, and “calories understood in no time”. This is a link to the URL to be executed.

  Thus, the information management server 10 can suitably extend the seed directory by the added URL and click directory.

<Fourth embodiment>
Next, a fourth embodiment of the present invention will be described.

[Functional configuration of information management server 10]
FIG. 27 is a functional block diagram of the information management server 10 of this embodiment. The storage unit 110 of the information management server 10 includes a click data storage unit 111, a web directory storage unit 112, an access log storage unit 113, a selection criterion storage unit 114, a selection criterion storage unit 115, and a subdirectory determination criterion. A storage unit 116.

Here, in the fourth embodiment, a directory to be personalized (for example, corresponding to a subdirectory D51 described later) is particularly referred to as a seed directory. The sea directory is created and stored in the web directory storage unit 112 in advance.
In the present embodiment, the information management server 10 is mainly intended to personalize the seed directory for each user who operates the terminal device 30.

  Here, the click data storage unit 111, the web directory storage unit 112, the access log storage unit 113, the selection criterion storage unit 114, and the selection criterion storage unit 115 are similar to the configuration of the storage unit 110 of the third embodiment described above. Therefore, the description is omitted.

  The subdirectory determination criterion storage unit 116 is used when determining whether or not to add the subdirectory created by the subdirectory creation unit 158 to the seed directory. For example, the conditions A to D shown below can be used as the subdirectory determination criterion.

  (Condition A) The subdirectory creation unit 158 uses all the click URLs and queries stored in the click data storage unit 111 when creating a subdirectory (click directory).

  (Condition B) When the subdirectory creation unit 158 creates a subdirectory (click directory), the click URL and query stored in the click data storage unit 111 are accessed by the terminal device 30 operated by the personalizing user. A sub-directory is created using only the URL containing the specified URL.

(Condition C) When creating the subdirectory (click directory), the subdirectory creation unit 158 uses only the one including the query input by the terminal device 30 operated by the user to be personalized, and creates the subdirectory. create.
Under conditions A to C, the technique described in the second embodiment is used when creating a subdirectory. Specifically, the similarity is calculated using a method of calculating the similarity between the virtual directory and the existing directory (seed directory). Then, the calculated similarity is determined, and a subdirectory is created according to the determination result.

(Condition D) Conditions A to D are mixed at a predetermined mixing ratio. For example, the mixing ratio can be set to 0.4 for condition A, 0.2 for condition B, and 0.4 for condition 6 where the entire mixing ratio of conditions A to D is 1.
In this case, the similarity for each condition is multiplied by the mixing ratio for each condition, and the multiplied value is added. Then, it is possible to determine whether or not to create a subdirectory using the added value.

  The control unit 120 includes a link creation unit 153, a transition probability calculation unit 154, a threshold determination unit 155, a subdirectory creation unit 158, a subdirectory determination unit 159, a subdirectory addition unit 160, and a selection selection unit 161. Is provided.

  Here, the link creation unit 153, the transition probability calculation unit 154, the threshold determination unit 155, and the selection selection unit 157 of the third embodiment are respectively the link creation unit 153, the transition probability calculation unit 154, and the threshold determination of the fourth embodiment. Since the functional configuration is the same as that of the unit 155 and the selection unit 161, the description thereof is omitted.

  When the third transition probability P (q | Dir) is larger than the second threshold as a result of the determination by the threshold determination unit 155, the subdirectory creation unit 158 determines the first directional link and the second directional link. A sub-directory (SubDir) to which the name of the query (q) that constitutes is assigned is created. The subdirectory thus created is also called a click directory. That is, the subdirectory creation unit 158 creates a click directory as a subdirectory in the seed directory.

  The subdirectory determination unit 159 determines whether or not to add the subdirectory created by the subdirectory creation unit 158 to the seed directory in order to personalize the seed directory based on the subdirectory determination criterion.

  The subdirectory adding unit 160 adds the subdirectory created by the subdirectory creating unit 158 to the seed directory according to the determination result by the subdirectory determining unit 159. Specifically, the subdirectory adding unit 160 adds the created subdirectory to the seed directory when the subdirectory created by the subdirectory creating unit 158 satisfies the subdirectory determination criterion.

  As described above, the information management server 10 according to this embodiment adds a subdirectory personalized to the user who operates the terminal device 30 to the seed directory based on the URL and the input query stored in the storage unit. Can do. Therefore, the information management server 10 can personalize the directory without requiring an operation from the user.

  In addition, the information management server 10 selects each URL stored in the added subdirectory. For this reason, the information management server 10 can maintain the URL stored in the subdirectory in a state suitable for a search in a specific specialized field.

[Process flow]
Next, the flow of processing of the information management server 10 of this embodiment will be described with reference to the flowcharts of FIGS. FIGS. 30 and 31 are schematic diagrams illustrating processing executed by the information management server 10 of the present embodiment for the web directory W3.

  Step S61: The link creation unit 153, transition probability calculation unit 154, threshold determination unit 155, and subdirectory creation unit 158 execute click directory creation processing.

  Step S62: The subdirectory determination unit 159 determines whether or not to add the subdirectory created by the subdirectory creation unit 158 in step S61 to the seed directory based on the subdirectory determination criterion. When a subdirectory is added to the seed directory (Yes), the process proceeds to step S63. On the other hand, when the subdirectory is not added to the seed directory (No), the series of processes is terminated.

  Step S63: The subdirectory adding unit 160 creates the subdirectory created in the seed directory by the subdirectory creating unit 158 according to the determination result that the subdirectory determining unit 159 determines to add the subdirectory to the seed directory in Step S62. Add Specifically, as shown in FIG. 31, URL 61, URL 62, and URL 63 are added in the seed directory D61 (see FIG. 30) before the URL is added.

  Step S64: The selection selection unit 161 is stored in the click directory added to the seed directory D31 based on the access log stored in the access log storage unit 113 or the selection criterion stored in the selection criterion storage unit 115. Each URL (link) is adjusted, and a series of processing is terminated.

  FIG. 29 is a flowchart showing the flow of click directory creation processing. Here, the click directory creation process shown in the flowchart of FIG. 29 is the same process as the click directory creation process of the third embodiment, and thus the description thereof is omitted.

[Specific example of web directory W3]
Next, a specific example of the web directory W3 according to the present embodiment will be described with reference to FIGS.
32 to 34 are diagrams showing specific display examples of the web directory W3. As shown in FIG. 32, the display screen 70 of the web directory W3 includes each seed directory 71 constituting the web directory W3, and a registration site 72 that is a URL stored (linked) in the seed directory 71. A search word input field 73 for inputting a query and a search button 74 are included.

  FIG. 33 is a diagram showing a specific example of the click directory (subdirectory) 75 created and added to the seed directory 71 in the seed directory 71 shown in FIG. As shown in FIG. 33, the click directory 75 is assigned category names of “summer recipe”, “recommended dish”, “shogayaki”, and “western recipe”, respectively.

  FIG. 34 is a diagram showing URLs stored in the click directory 75a of the category name “Summer Recipe” in the click directory 66 shown in FIG. As shown in FIG. 34, the URL 76 stored in the click directory 75a of the category name “Summer Recipe” is “Delicious in Summer! Summer Cooking Recipe”, “Summer! It is a link to a URL to which the names “summer recipe” and “rainy season recipe” are given.

  Thus, the information management server 10 can expand the seed directory for the user who operates the terminal device 30.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

It is a lineblock diagram showing search system 1 concerning one embodiment of an information management device of the present invention. It is a figure which shows an example of the hardware constitutions of the information management server 10 which concerns on one Embodiment of the information management apparatus of this invention. It is a functional block diagram of the information management server 10 of 1st Embodiment. It is a figure which shows an example of the click data table 111a. 3 is a schematic diagram showing a web directory W stored in a web directory storage unit 112. FIG. It is a flowchart explaining the flow of a process of the information management server 10 of 1st Embodiment. It is a flowchart explaining the flow of a process of the information management server 10 of 1st Embodiment. It is a flowchart explaining the flow of a process of the information management server 10 of 1st Embodiment. It is a figure shown about the directed link created by the link creation part. It is a figure shown about the directed link created by the link creation part. 3 is a schematic diagram showing a web directory W stored in a web directory storage unit 112. FIG. 3 is a schematic diagram showing a web directory W stored in a web directory storage unit 112. FIG. 3 is a schematic diagram showing a web directory W stored in a web directory storage unit 112. FIG. It is a functional block diagram of the information management server 10 of 2nd Embodiment. It is a flowchart explaining the flow of a process of the information management server 10 of 2nd Embodiment. It is a functional block diagram of the information management server 10 of 3rd Embodiment. 3 is a schematic diagram showing a web directory W2 stored in a web directory storage unit 112. FIG. It is a flowchart explaining the flow of a process of the information management server 10 of 3rd Embodiment. It is a flowchart shown about the flow of a click directory creation process. 3 is a schematic diagram showing a web directory W2 stored in a web directory storage unit 112. FIG. 3 is a schematic diagram showing a web directory W2 stored in a web directory storage unit 112. FIG. 3 is a schematic diagram showing a web directory W2 stored in a web directory storage unit 112. FIG. It is a figure shown about the specific example of a display of the web directory W2. It is a figure shown about the specific example of a display of the web directory W2. It is a figure shown about the specific example of a display of the web directory W2. It is a figure shown about the specific example of a display of the web directory W2. It is a functional block diagram of the information management server 10 of 4th Embodiment. It is a flowchart explaining the flow of a process of the information management server 10 of 4th Embodiment. It is a flowchart shown about the flow of a click directory creation process. 3 is a schematic diagram showing a web directory W3 stored in a web directory storage unit 112. FIG. 3 is a schematic diagram showing a web directory W3 stored in a web directory storage unit 112. FIG. It is a figure shown about the specific example of a display of the web directory W3. It is a figure shown about the specific example of a display of the web directory W3. It is a figure shown about the specific example of a display of the web directory W3.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Search system 10 Information management server 20 Search server 30 Terminal device 40 Content server 50 Communication line 110 Memory | storage part 111 Click data storage part 120 Control part

Claims (21)

In an information management device that manages a URL stored in a directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query,
Storage means for storing a URL clicked by the user and an input query input by the user terminal among URLs of search results obtained by searching the search query by the search engine;
Based on the URL stored in the storage unit and the input query, a first directional link is created from the URL stored in the directory to the input query for obtaining the URL as the search result. Link creation means;
Based on the URL and the input query stored in the storage means, the input query is selected from the search results URL searched by the search engine from the input query constituting the first directed link. Second link creating means for creating a second directed link to the selected URL;
Transition probability calculating means for calculating a transition probability from the directory in which the URL constituting the first directional link is stored to the selected URL constituting the second directional link;
First threshold determination means for determining whether or not the transition probability is greater than a first threshold;
URL determination means for determining whether or not the selected URL exists in the directory when the transition probability is larger than the first threshold as a result of the determination by the first threshold determination means;
An information management apparatus comprising: a display order changing means for changing the display order of the selected URL in the directory when the selected URL exists in the directory as a result of the determination by the URL determining means.   2. The information according to claim 1, further comprising URL adding means for adding the selected URL to the directory when the selected URL does not exist in the directory as a result of the determination by the URL determining means. Management device. The transition probability calculation means includes
Based on the URL and the input query stored in the storage means, a first transition for calculating a first transition probability from a URL constituting the first directional link to a query constituting the first directional link Probability calculation means;
Based on the URL and the input query stored in the storage means, a second from a query constituting the first directional link and the second directional link to a selected URL constituting the second directional link. A second transition probability calculating means for calculating a transition probability;
Based on the first transition probability and the second transition probability, the transition probability from the directory storing the URL constituting the first directional link to the selected URL constituting the second directional link. The information management apparatus according to claim 1, further comprising third transition probability calculation means for calculating A fourth transition probability calculating means for calculating a fourth transition probability from the directory to an input query constituting the first directional link and the second directional link based on the first transition probability;
Second threshold determination means for determining whether or not the fourth transition probability is greater than a second threshold;
As a result of the determination, if the fourth transition probability is greater than a second threshold, the names of the input queries constituting the first directional link and the second directional link are assigned to the directory. The information management apparatus according to claim 3, further comprising subdirectory creation means for creating a subdirectory. Third threshold value determining means for determining whether or not the transition probability is greater than a third threshold value;
If the transition threshold is larger than the third threshold as a result of the determination by the third threshold determination means, the apparatus further comprises subdirectory URL addition means for adding the selected URL to the created subdirectory. The information management apparatus according to claim 4. Similarity calculation means for calculating the similarity between the directory and the created subdirectory;
Similarity determination means for determining whether or not the calculated similarity satisfies a predetermined condition;
6. The apparatus according to claim 5, further comprising: a link creating unit that creates a link between the directory and the created subdirectory when the degree of similarity satisfies the predetermined condition as a result of the determination. The information management device described in 1. In an information management device that manages a URL stored in a directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query,
Storage means for storing a URL clicked by the user and an input query input by the user terminal among URLs of search results obtained by searching the search query by the search engine;
A frequency calculating means for calculating a frequency at which a URL stored in the directory is selected based on the URL and the input query stored in the storage means;
URL extraction means for extracting a URL that exceeds a predetermined frequency among the calculated frequencies;
A degree-of-similarity calculation means for calculating the degree of similarity with the directory, using the set of extracted URLs as a virtual directory;
Similarity determination means for determining whether the calculated similarity satisfies a predetermined condition;
If the predetermined condition is satisfied as a result of the determination, the information further comprises subdirectory creation means for creating a subdirectory in which the extracted URL is stored in the directory where the similarity is calculated. Management device.   As a result of the determination, when the predetermined condition is satisfied, the extracted URL is added to the subdirectory, and a URL that overlaps with the added URL and the URL stored in the directory where the similarity is calculated. The information management apparatus according to claim 7, further comprising URL merging means for merging. A URL stored in a directory, a search engine for searching a web page, and a user terminal for a user to input a search query, wherein the search query is a URL of a search result searched by the search engine, In an information management method executed by an information management apparatus in which a URL clicked by a user and an input query input by the user terminal are stored in association with each other,
A first link creating step for creating a first directional link to the input query for obtaining the URL as the search result from the URL stored in the directory based on the stored URL and the input query When,
Based on the stored URL and the input query, the selected URL selected from the URLs of the search results in which the input query is searched by the search engine from the input queries constituting the first directed link. A second link creating step for creating a second directed link to
A transition probability calculating step of calculating a transition probability from the directory in which the URL configuring the first directional link is stored to the selected URL configuring the second directional link;
A threshold determination step of determining whether or not the transition probability is greater than a predetermined threshold;
As a result of the determination by the threshold determination step, if the transition probability is greater than the predetermined threshold, a URL determination step of determining whether or not the selected URL exists in the directory;
An information management method comprising: a display order changing step of changing a display order of the selected URL in the directory when the selected URL exists in the directory as a result of the determination in the URL determining step. A URL stored in a directory, a search engine for searching a web page, and a user terminal for a user to input a search query, wherein the search query is a URL of a search result searched by the search engine, A computer program executed by a computer of an information management apparatus in which a URL clicked by a user and an input query input by the user terminal are stored in association with each other,
A first link creating step for creating a first directional link to the input query for obtaining the URL as the search result from the URL stored in the directory based on the stored URL and the input query When,
Based on the stored URL and the input query, the selected URL selected from the URLs of the search results in which the input query is searched by the search engine from the input queries constituting the first directed link. A second link creating step for creating a second directed link to
A transition probability calculating step of calculating a transition probability from the directory in which the URL configuring the first directional link is stored to the selected URL configuring the second directional link;
A threshold determination step of determining whether or not the transition probability is greater than a predetermined threshold;
As a result of the determination by the threshold determination step, if the transition probability is greater than the predetermined threshold, a URL determination step of determining whether or not the selected URL exists in the directory;
A computer program for causing a computer to execute a display order changing step for changing the display order of the selected URL in the directory when the selected URL exists in the directory as a result of the determination in the URL determining step. Manage URLs stored in seed directories for specific fields of expertise, search engines that search web pages, and user terminals from which users enter search queries, and specialize the seed directory to the specific fields of expertise An information management device,
Storage means for storing a URL clicked by the user and an input query input by the user terminal among URLs of search results obtained by searching the search query by the search engine;
Of the URL and the input query stored in the storage unit, the URL is related to the seed directory based on a first selection criterion from the URL and the input query by all users stored in the storage unit. First extraction means for extracting the URL and the input query;
Second extraction means for extracting a URL to be added to the seed directory based on a second selection criterion from the URL extracted by the first extraction means;
An information management apparatus comprising: URL addition means for adding the URL extracted by the second extraction means to the seed directory. Based on the URL and the input query stored in the storage means, a first directional link to the input query for obtaining the URL as the search result from the URL stored in the seed directory is generated. 1 link creation means,
Based on the URL and the input query stored in the storage means, the input query is selected from the search query URLs searched by the search engine from the input queries constituting the first directed link. Second link creating means for creating a second directed link to the selected URL;
Transition probability calculating means for calculating a transition probability from the seed directory storing the URL constituting the first directional link to the selected URL constituting the second directional link;
Based on the transition probability calculated by the transition probability calculating means, a subdirectory for creating a subdirectory in which the names of input queries constituting the first directional link and the second directional link are assigned to the seed directory Directory creation means;
12. The information management apparatus according to claim 11, further comprising subdirectory addition means for adding the subdirectory created by the subdirectory creation means to the seed directory. The first link creating means is input by the user terminal accessing the URL and the seed directory entered by the user accessing the seed directory among the URL and the input query stored in the storage means. Creating a first directional link to the input query for obtaining the URL as the search result from the URL stored in the seed directory based on the input query
The second link creating means is input by the user terminal that has accessed the URL and the seed directory that are input by the user that has accessed the seed directory, among the URL and the input query that are stored in the storage means. Based on the input query that has been made, the second query from the input query that constitutes the first directional link to the selected URL that is selected from among the search result URLs in which the input query was searched by the search engine. The information management apparatus according to claim 12, wherein a direction link is created.   13. The information management apparatus according to claim 12, further comprising a sorting unit that sorts each URL stored in the subdirectory added by the subdirectory adding unit. A search result obtained by managing a URL stored in a seed directory related to a specific specialized field, a search engine that searches a web page, and a user terminal from which a user inputs a search query, wherein the search query is searched by the search engine The URL clicked by the user and the input query input by the user terminal are stored in association with each other, and are executed by an information management device for specializing the seed directory in the specific specialized field. Information management method,
Of the stored URL and the input query, the URL and the input query related to the seed directory based on a first selection criterion from the stored URL and the input query by all users. A first extraction step of extracting
A second extraction step of extracting a URL to be added to the seed directory based on a second selection criterion from the URL extracted in the first extraction step;
A URL adding step of adding the URL extracted in the second extraction step to the seed directory. A search result obtained by managing a URL stored in a seed directory related to a specific specialized field, a search engine that searches a web page, and a user terminal from which a user inputs a search query, wherein the search query is searched by the search engine The URL clicked by the user and the input query input by the user terminal are stored in association with each other, and are executed by a computer of an information management apparatus that specializes the seed directory in the specific specialized field. A computer program,
Of the stored URL and the input query, the URL and the input query related to the seed directory based on a first selection criterion from the stored URL and the input query by all users. A first extraction step of extracting
A second extraction step of extracting a URL to be added to the seed directory based on a second selection criterion from the URL extracted in the first extraction step;
A computer program for causing a computer to execute a URL adding step of adding the URL extracted in the second extracting step to the seed directory. An information management apparatus that manages a URL stored in a seed directory, a search engine that searches a web page, and a user terminal from which a user inputs a search query, and personalizes the seed directory to the user. ,
Storage means for storing a URL clicked by the user and an input query input by the user terminal among URLs of search results obtained by searching the search query by the search engine;
Of the URL and the input query stored in the storage unit, based on the URL by all users and the input query by all users stored in the storage unit, from the URL stored in the seed directory First link creating means for creating a first directed link to the input query for obtaining a URL as the search result;
Of the URL and the input query stored in the storage unit, the first directional link is configured based on the URL by all users and the input query by all users stored in the storage unit A second link creating means for creating a second directional link from the input query to the selected URL selected from the URLs of search results obtained by searching the input query by the search engine;
Transition probability calculating means for calculating a transition probability from the seed directory storing the URL constituting the first directional link to the selected URL constituting the second directional link;
Based on the transition probability calculated by the transition probability calculating means, a subdirectory for creating a subdirectory in which the names of input queries constituting the first directional link and the second directional link are assigned to the seed directory Directory creation means;
Subdirectory determination means for determining whether to add the subdirectory created by the subdirectory creation means to the seed directory based on a subdirectory determination criterion;
An information management apparatus comprising: a subdirectory adding unit that adds a subdirectory created by the subdirectory creating unit to the seed directory in accordance with a determination result by the subdirectory determining unit. The first link creating means is input by the user terminal accessing the URL and the seed directory entered by the user accessing the seed directory among the URL and the input query stored in the storage means. Creating a first directional link to the input query for obtaining the URL as the search result from the URL stored in the seed directory based on the input query
The second link creating means is input by the user terminal that has accessed the URL and the seed directory that are input by the user that has accessed the seed directory, among the URL and the input query that are stored in the storage means. Based on the input query that has been made, the second query from the input query that constitutes the first directional link to the selected URL that is selected from among the search result URLs in which the input query was searched by the search engine. The information management apparatus according to claim 17, wherein a direction link is created.   The information management apparatus according to claim 17, further comprising a sorting unit that sorts each URL stored in the subdirectory added by the subdirectory adding unit. A URL stored in the seed directory, a search engine that searches a web page, and a user terminal that a user inputs a search query are managed. Among the URLs of search results that are searched by the search engine, An information management method executed by an information management apparatus that stores a URL clicked by the user and an input query input by the user terminal in association with each other and personalizes the seed directory to the user,
Of the stored URL and the input query, the URL is obtained as the search result from the URL stored in the seed directory based on the stored URL and the input query by all users. Creating a first directed link to the input query for creating a first directed link;
Of the stored URL and the input query, based on the stored URL and the input query by all users, the input query from the input query constituting the first directed link A second link creating step of creating a second directional link to the selected URL selected from the search result URLs searched by the search engine;
A transition probability calculating step of calculating a transition probability from the seed directory in which the URL constituting the first directional link is stored to the selected URL constituting the second directional link;
Based on the transition probability calculated by the transition probability calculation step, a subdirectory for creating a subdirectory in which the names of input queries constituting the first directional link and the second directional link are assigned to the seed directory A directory creation step;
A subdirectory determination step for determining whether or not to add the subdirectory created by the subdirectory creation step to the seed directory based on a subdirectory determination criterion;
An information management method comprising: a subdirectory addition step of adding the subdirectory created in the subdirectory creation step to the seed directory in accordance with a determination result in the subdirectory determination step. A URL stored in the seed directory, a search engine that searches a web page, and a user terminal that a user inputs a search query are managed. Among the URLs of search results that are searched by the search engine, A computer program executed by a computer of an information management apparatus that stores a URL clicked by the user and an input query input by the user terminal in association with each other and personalizes the seed directory to the user. ,
Of the stored URL and the input query, the URL is obtained as the search result from the URL stored in the seed directory based on the stored URL and the input query by all users. Creating a first directed link to the input query for creating a first directed link;
Based on the stored URL and the input query by all users among the stored URL and the input query, the input query from the input query constituting the first directed link is A second link creating step of creating a second directional link to the selected URL selected from the search result URLs searched by the search engine;
A transition probability calculating step of calculating a transition probability from the seed directory in which the URL constituting the first directional link is stored to the selected URL constituting the second directional link;
Based on the transition probability calculated by the transition probability calculation step, a subdirectory for creating a subdirectory in which the names of input queries constituting the first directional link and the second directional link are assigned to the seed directory A directory creation step;
A subdirectory determination step for determining whether or not to add the subdirectory created by the subdirectory creation step to the seed directory based on a subdirectory determination criterion;
A computer program that causes a computer to execute a subdirectory addition step of adding a subdirectory created in the subdirectory creation step to the seed directory in accordance with a determination result in the subdirectory determination step.

Images