JP2009265754A - Information providing system, information providing method, and information providing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information providing system, an information providing method, and an information providing program can easily provide information useful for a user with a high degree of precision. <P>SOLUTION: A user similarity degree calculation part 11 calculates a similarity degree among users indicating a similarity of the trend of access to a content among users from the information on the frequency of access to each content by the users. A content score calculation part 12 calculates a content score, which is the information indicating the usefulness of a content for users from the information on the frequency of access by each user which is weighted by the similarity degree among the users, and stores it as an index. A provision information generation part 15 refers to the index, and obatains content identification information that matches a query transmitted from a communication terminal device 30, and a provision information output part 16 outputs the obtained content identification information to the communication terminal device 30 as provision information. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an information providing apparatus, an information providing method, and an information providing program for transmitting content identification information of content that matches a query transmitted from a communication terminal apparatus to the communication terminal apparatus.

  Currently, a large amount of information is stored on the Internet, and a user accesses desired information by a method such as keyword search. In a conventional search engine, a document that matches a condition is extracted using a relationship between a search condition input by a user and document information such as a keyword distribution of a stored document group. It is common to provide a user with However, since there is a difference in interest and preference for each user, the usefulness of each piece of information differs for each user even if the information is extracted from the same search condition. For this reason, presentation of information reflecting user-specific information is required.

  Patent Document 1 describes a system in which attribute information such as a user's age, gender, and address is registered in advance as a user profile, and contents such as advertisements are presented to the user based on the user profile. Each content is associated with attribute information of the user to be appealed by an information sender, etc., and the content to be presented to the user is determined by comparing this attribute information with the user profile.

  In Patent Document 2, based on the information access history of a large number of users, a related score between a set of combinations of search keywords and user attribute information and content that is a candidate for presentation is calculated in advance and held as a search index. A device that extracts a search keyword input by a user and content associated with the user attribute information of the user with reference to a search index and presents the content in the order of related scores is described.

  Patent Literature 3 records a search keyword and a related word input by a user in a document search, calculates a similarity between users from a related word input history, and stores a related word list of other users having a high similarity. Describes a document retrieval device to be presented. According to this document search apparatus, input of related terms for preventing a search omission is supported, and an efficient document search is possible.

Patent Document 4 discloses information that calculates the degree of similarity between the user and the other user based on the URL access history of the user and the other user, and provides the user with information such as the access history of the other user with a high degree of similarity. A processing device is described. According to this information processing apparatus, a user can access a document used by another user who has a high probability of having a common hobby and preference.
JP-A-10-134080 US Pat. No. 6,539,377 JP 2007-148948 A JP 2007-141135 A

  In the systems described in Patent Literature 1 and Patent Literature 2, since user attribute information is used, similar information is presented to a plurality of users having the same attribute even if the content preference is different. Is done. That is, there is a problem in terms of usefulness of information presented to the user. In addition, since user attribute information is manually input, there is a problem that the burden of this input is large. Furthermore, there is a problem that the accuracy of the attribute information is low because attribute information that is different from the fact is intentionally input or the user's attribute changes over time. In addition, since the search keyword is selected manually, even if the same information is requested, there is a difference in the ranking of the content presented by the input search keyword, and there is a problem in the accuracy of the presented information.

  In addition, in the devices described in Patent Document 3 and Patent Document 4, information associated with other users having a high degree of similarity to the user remains as it is for the user without considering the weight of each information. Was presented. Therefore, information useful for the user has not always been preferentially presented. That is, there is a problem that accuracy is low in terms of usefulness of the presented information.

  SUMMARY An advantage of some aspects of the invention is that it provides an information providing apparatus, an information providing method, and an information providing program that can provide information useful to a user accurately and easily.

  In order to solve the above problems, an information providing apparatus according to the present invention includes an access history storage unit that stores access frequency information indicating a frequency of access to each content by a user in association with user identification information for identifying the user. An inter-user similarity calculating means for calculating the similarity between users representing the similarity of access tendency to content among the users based on the access frequency information stored in the access history storage means, and the user and each user The content score calculation means for calculating the content score, which is the information representing the usefulness of the content for the user, from the access frequency information of each user weighted by the similarity between the users between, and the content score calculation means Corresponding content score of each content to user identification information Index storage means for storing, query input means for receiving an input of a query including at least user identification information transmitted from the communication terminal device, and content identification information of content that conforms to the query accepted by the query input means , Provision information generation means for generating provision information from the acquired content identification information with reference to the index storage means, and provision information output means for outputting the provision information generated by the provision information generation means to the communication terminal device It is characterized by that.

  In order to solve the above-described problem, the information providing method of the present invention is the information providing method in the information providing apparatus, in which the information providing apparatus stores the access frequency information indicating the frequency of access to each content by the user. Similarity between users that calculates similarity between users representing similarity of access tendency to content among users based on access frequency information stored in access history storage means stored in association with identifying user identification information The content score, which is information representing the usefulness of content for the user, is calculated from the degree calculation step and the access frequency information of each user weighted by the similarity between users between the user and each user. Stores the content score of each content as an index in association with user identification information A content score calculation step to be stored in a stage, a query input step for receiving an input of a query including at least user identification information transmitted from the communication terminal device, and content identification information for content matching the query received by the query input step Obtaining and referring to the index storage means, a provision information generating step for generating provision information from the acquired content identification information, and a provision information output step for outputting the provision information generated in the provision information generation step to the communication terminal device It is characterized by performing.

  Furthermore, in order to solve the above problem, the information providing program of the present invention stores the computer in association with the user identification information for identifying the user, the access frequency information indicating the frequency of access to each content by the user. An access history storage means, and an inter-user similarity calculation means for calculating the similarity between users representing the similarity of the access tendency to the content among the users based on the access frequency information stored in the access history storage means; A content score calculating means for calculating a content score, which is information representing the usefulness of content for the user, from the access frequency information of each user weighted by the similarity between users between the user and each user; and a content score Content contents of each content calculated by the calculation means Conforms to the index storage means for storing the user ID in association with the user identification information, the query input means for receiving the input of the query including at least the user identification information transmitted from the communication terminal device, and the query received by the query input means The content identification information of the content to be acquired is acquired, the provision information generation means for generating the provision information from the acquired content identification information with reference to the index storage means, and the provision information generated by the provision information generation means to the communication terminal device It is characterized by functioning as provision information output means for outputting.

  With these configurations, the similarity between users is calculated based on the similarity of the access tendency to each content. The content score calculation means extracts the access frequency of each user for each content from the access history, and calculates the content score from the access frequency weighted by the similarity between users between the users. For this reason, since the content score of the content accessed by the other user increases as the similarity between the user and the other user increases, the content score accurately represents the usefulness of the document to the user. it can. Since the content identification information of the content that matches the query is presented to the user based on the content score, a document that is highly useful to the user is presented. Further, the content score is calculated from the user's access history with respect to the content, and it is not necessary to input special attribute information or the like, so the burden on the user is small. Therefore, it is possible to provide information useful for the user accurately and easily.

  In the information providing apparatus according to the present invention, the provided information generating unit refers to the index storing unit and the access history storing unit, and the content specified by the user identification information included in the query has not been accessed by the user. The provision information can be generated only from the content identification information.

  With these configurations, content that the user has never accessed is extracted based on the user identification information included in the query, and the extracted content is presented to the user based on the content score. Accordingly, since the content unknown to the user is presented, it is possible to provide information more useful to the user.

  An information providing apparatus according to the present invention collects an access history for each content by a user, generates access frequency information based on the collected access history, and stores the generated access frequency information in an access history storage unit It further comprises a collecting means.

  With this configuration, since the access frequency information is generated based on the history of access to the content, the access frequency information is easily generated, and the present invention can be reliably implemented.

  In the information providing apparatus according to the present invention, the access history collecting means collects the access history of access to the content output from the communication terminal device in response to the provided information, and is included in the provided information and collects the access history. It is characterized in that the access frequency information of the content that has not been reduced is reduced.

  With this configuration, it is possible to collect a history of accesses performed according to the provided information presented to the user, so that content that the user did not access despite being presented to the user as provided information is specified. be able to. By reducing the access frequency information of the content that has not been accessed in this way, the usefulness of the content for the user is more strongly reflected in the access frequency information. By calculating the content score using such access frequency information, it is possible to further improve the accuracy of the usefulness of the provided content for the user.

  In the information providing apparatus according to the present invention, the provided information generating unit generates the provided information in which the content identification information is arranged in the order of the content score with reference to the index storing unit.

  With this configuration, content identification information having a large content score is presented to the user, so that it is possible to provide the user with highly useful information for the user.

  In the information providing apparatus according to the present invention, the query transmitted from the communication terminal apparatus includes a search keyword, and the provided information generation unit acquires content identification information of content including the search keyword.

  With this configuration, the information providing apparatus of the present invention can be applied to document search using keywords.

  In the information providing apparatus according to the present invention, the content identification information can include information indicating the location of the content on the network.

  With this configuration, based on the provided content identification information, the user can access content located on the network.

  In the information providing apparatus according to the present invention, it is preferable that the content score calculating means calculates a content score using an access history of a user whose similarity between users is higher than a predetermined threshold.

  In the information providing apparatus of the present invention, the number of accessing users is generally very large. When calculating the content score, it is possible to exclude access histories of other users whose similarity between users is equal to or less than a predetermined threshold value, so that the processing load can be reduced.

  According to the information providing apparatus, the information providing method, and the information providing program of the present invention, the content identification information of a plurality of contents that match the query is presented to the user based on the content score representing the usefulness of the content for the user. Therefore, content that is highly useful to the user is presented. Since the content score is calculated from the access history, there is no burden such as attribute input by the user. As a result, it is possible to provide information useful for the user accurately and easily.

DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals.
[First Embodiment]

  FIG. 1 is a block diagram illustrating functional configurations of the information providing apparatus 10, the network access server 20, and the communication terminal apparatus 30 according to the first embodiment. The information providing apparatus 10 and the communication terminal apparatus 30 can communicate with each other via a wired or wireless network 40. The network 40 is, for example, a mobile communication network or the Internet.

  The information providing device 10 is a device that acquires content identification information of content that conforms to the query transmitted from the communication terminal device 30, and transmits the acquired content identification information to the communication terminal device 30, and is configured by a server computer, for example. The The content is, for example, an html file, a text file, a rich text file, an image file, or the like. The content identification information can include information (for example, URL) indicating the location of the content on the network, for example.

  The network access server 20 is a device that controls access from the communication terminal device 30 to various devices on the network 40, and is, for example, a gateway or a proxy.

  The communication terminal device 30 is a device that transmits / receives information to / from the information providing device 10 and the network access server 20, and includes, for example, a mobile terminal device and a personal computer. The communication terminal device 30 transmits a query to the information providing device 10, displays the provided information transmitted from the information providing device 10 in response to the query, and presents it to the user. A plurality of users can share one communication terminal device 30 or a single user can use a plurality of communication terminal devices 30, but in the first embodiment, the description For convenience, it is assumed that one user uses one communication terminal device 30. Further, the communication terminal device 30 may communicate with the information providing device 10 via the network access server 20. Here, although the three communication terminal devices 30 are shown in FIG. 1, it is not restricted to this number.

  The functional elements of the information providing apparatus 10 and the network access server 20 will be described together using FIG. Details of processing performed in each functional element will be described later. In the first embodiment, the information providing apparatus 10 and the network access server 20 are configured as separate apparatuses, but may be configured as one apparatus. Functionally, the information providing apparatus 10 includes an inter-user similarity calculation unit 11, a content score calculation unit 12, an index storage unit 13, a query input unit 14, a provision information generation unit 15, and a provision information output unit 16. . Functionally, the network access server 20 includes an access history collection unit 21 and an access history storage unit 22.

  The access history collection unit 21 collects a history of access to each content by a user, generates access frequency information based on the collected access history, and stores the generated access frequency information in the access history storage unit 22 Means. This access history is information indicating who accessed what content, and includes at least user identification information and content identification information, and can further include time information, information for identifying the communication terminal device 30, and the like. The access history collection unit 21 uses access destination information (for example, URL) based on access performed from the communication terminal apparatus 30 via the network access server 20, and information (for example, IP address) for identifying the communication terminal apparatus 30. And terminal identification numbers such as IMEI), user identification information (for example, login name at the time of network access, SIM or UIM contractor identification number), and time information are collected as information indicating an access history. These pieces of information may be directly transmitted from the communication terminal device 30.

  The access history storage unit 22 is an access history storage unit that stores access frequency information indicating the frequency of access to each content by a user in association with user identification information for identifying the user.

  The inter-user similarity calculation unit 11 calculates the inter-user similarity indicating the similarity of the access tendency to the content among the users based on the access frequency information stored in the access history storage unit 22. It is a calculation means. The inter-user similarity calculation unit 11 sends the calculated inter-user similarity to the content score calculation unit 12.

  The content score calculation unit 12 calculates a content score, which is information representing the usefulness of content for the user, from the access frequency information of each user weighted by the similarity between users between the user and each user. It is a score calculation means. Here, each user includes the user and other users. Further, the content score calculation unit 12 causes the index storage unit 13 to store information in which the content score of each content calculated for each user is associated with the user identification information as an index.

  The index storage unit 13 is an index storage unit that stores an index in which a content score of each content is associated with user identification information.

  The query input unit 14 is a query input unit that receives an input of a query including at least user identification information transmitted from the communication terminal device 30. The query is a request for providing information to the information providing apparatus 10 and can include a search keyword and other search conditions. The query input unit 14 sends the accepted query to the provision information generation unit 15.

  The provision information generation unit 15 acquires content identification information of content that conforms to the query received by the query input unit, and refers to the index storage unit 13 to generate provision information from the acquired content identification information Means. The provision information generation unit 15 sends the generated provision information to the provision information output unit 16.

  The provision information output unit 16 is provision information output means for outputting the acquired provision information to the communication terminal device 30.

  FIG. 2 is a hardware configuration diagram of the information providing apparatus 10. The information providing apparatus 10 physically includes a CPU 101, a RAM 102 and a ROM 103 which are main storage devices, a communication module 104 which is a data transmission / reception device such as a hard disk and a network card, an auxiliary storage device 105 such as a flash memory, and the like. It is configured as. The functions of the information providing apparatus 10 described in FIG. 1 operate the communication module 104 and the like under the control of the CPU 101 by reading predetermined computer software on the hardware such as the CPU 101 and the RAM 102 shown in FIG. In addition, it is realized by reading and writing data in the RAM 102 and the auxiliary storage device 105. The information providing apparatus 10 is realized by a server computer, for example. Although not shown, the network access server 20 also physically has a hardware configuration as shown in FIG.

  Subsequently, each functional element of the communication terminal device 30 will be described. The communication terminal device 30 includes an instruction receiving unit 31, a user identification information storage unit 32, a transmission / reception unit 33, and a display unit 34.

  The instruction receiving unit 31 is an instruction receiving unit that receives an instruction from a user for receiving information provided from the information providing apparatus 10.

  The user identification information storage unit 32 is a user identification information storage unit that stores user identification information for uniquely identifying a user. In this user identification information storage unit 32, user identification information of a user who is the owner of the communication terminal device 30 can be stored in advance. In addition, user identification information input when a login operation to the communication terminal device 30 is performed can be stored.

  The transmission / reception unit 33 is a transmission unit that transmits user identification information as a query to the information providing apparatus 10 when the instruction receiving unit 31 receives an instruction from the user, and receives the provision information transmitted from the information providing apparatus 10. Receiving means. The query can include a search keyword and other search conditions in addition to the user identification information. These search conditions are input by the user and received by the instruction receiving unit 31.

  The display unit 34 is a display unit that displays the received provision information, and is configured by, for example, a display device.

  FIG. 3 is a hardware configuration diagram of the communication terminal device 30. The communication terminal device 30 physically includes a CPU 301, a RAM 302 and a ROM 303 as main storage devices, an input device 304 such as a keyboard as an input device, a communication module 305 as a data transmission / reception device such as a network card, a hard disk, and a flash memory. The computer system includes an auxiliary storage device 306 such as a display and an output device 307 such as a display. Each function of the communication terminal device 30 described in FIG. 1 is obtained by reading predetermined computer software on the hardware such as the CPU 301 and the RAM 302 shown in FIG. 3, thereby controlling the communication module 305 and the input device under the control of the CPU 301. This is realized by operating 304, the output device 307, and the like, and reading and writing data in the RAM 302 and the auxiliary storage device 306.

  Subsequently, operations of the information providing apparatus 10, the network access server 20, and the communication terminal apparatus 30 according to the first embodiment will be described. FIG. 4 is a diagram illustrating a processing procedure in which the information providing apparatus 10 and the network access server 20 generate an index.

  First, when the communication terminal device 30 accesses content via the network access server 20, the access history collection unit 21 collects an access history and stores the collected access history in the access history storage unit 22 ( S101). An example of the data structure of the access history stored in the access history storage unit is shown in FIG. The access history is data in which a user ID that is user identification information is associated with a content ID that is content identification information. For example, the record in the first line of the access history shown in FIG. 5 indicates that the user U02 has accessed the content D04. Each time an access history is collected, data represented by a set of a user ID and a content ID is sequentially added and stored in the access history storage unit 22.

  Subsequently, the access history collection unit 21 of the network access server 20 converts the access history data illustrated in FIG. 5 into a data structure in which the frequency of accessing each content as illustrated in FIG. 6 is tabulated for each user. (S102). The converted data is stored in the access history storage unit 22. For example, the content D01 is shown to have been accessed five times by the user U01. In the first embodiment, the data structure conversion performed in step S102 is sequentially performed every time the access history is added to the access history storage unit 22, but the access history data shown in FIG. It may be performed when the predetermined amount is increased or at regular intervals. Further, the network access server 20 may store the access history with the data structure shown in FIG. 5 without performing the data structure conversion performed in step S102. In this case, the user similarity calculation unit 11 of the information providing apparatus 10 performs the data structure conversion shown in step S102.

  The inter-user similarity calculation unit 11 calculates the inter-user similarity based on the access history stored in the access history storage unit 22 (S103, inter-user similarity calculation step). This similarity between users is obtained by quantifying the similarity of access tendency to content. The calculation of the similarity between users can be realized by an existing technology. In the first embodiment, a vector having an access frequency for each content as an element is generated for each user, and an inner product value of vectors between users is set as a similarity between users. In this vector, the number of contents is the number of dimensions of the vector. Hereinafter, an example of the calculation of the similarity between users by the similarity calculation unit 11 between users will be specifically described.

First, the inter-user similarity calculation unit 11 normalizes the data shown in FIG. 6 so that the vector size is 1 with the number of accesses to each content for each user as the value of the vector element. For example, the value of the element indicating the frequency of access to the content D01 by the user U01 is obtained as 5 / (5 ² +2 ² +2 ² ) ^1/2 = 0.87. FIG. 7 shows data obtained by normalizing the access frequency for each content for each user. Furthermore, the similarity calculation part 11 between users arranges the data which normalized the access frequency with respect to each content for every user as an element value of a vector, arranges it in ascending order of content ID for every user, and produces | generates the user vector for every user To do. The generated user vector is shown in FIG. This user vector represents a user's access tendency to each content. In the present embodiment, for convenience of explanation, the number of elements N of the user vector is 4, but the value of the number N of elements is the total number of contents to be accessed. Become. The number of contents to be accessed may be the total number of contents accessed by any user or may include contents collected using a crawler that is an existing technology. However, when content collected using a crawler is targeted, content that has never been accessed by the user is included.

  Subsequently, the inter-user similarity calculation unit 11 calculates the inter-user similarity based on the user vector shown in FIG. The similarity between users is obtained by calculating an inner product value of user vectors. FIG. 9 shows the similarity between users obtained in this way. For example, the similarity between users U01 and U02 is 0.084. Since the similarity between users is an inner product value of vectors, the value is 0 or more and 1 or less. When the similarity between users is 0, it indicates that there is no access to the same content between users, and the similarity between users When the degree is 1, the normalized access frequency is the same. The inter-user similarity calculation unit 11 stores the generated inter-user similarity in a temporary storage unit (not shown). In addition, the similarity calculation part 11 between users is good also as calculating the similarity between users of a predetermined number or less number of users. In general, the information providing apparatus 10 provides a service that provides information to a huge number of users. Therefore, by limiting the calculation of the similarity between users to a predetermined number that does not deteriorate the accuracy of the provided information, The processing load on the information providing apparatus 10 can be reduced.

  Next, the content score calculation unit 12 calculates a content score for each user based on the similarity between users and the user vector generated in step S103 (S104, content score calculation step). The calculation of the content score of the user U02 will be specifically described below as an example. First, the content score calculation unit 12 calculates the similarity-considered user vector shown in FIG. 10 by multiplying the user vector of each user shown in FIG. 8 by the similarity between users U02 and each user. For example, when the user vector (0.87, 0.35, 0, 0.35) of the user U01 is multiplied by the similarity 0.084 between the users U02 and U01, the similarity of the user U01 to the user U02 A degree-considering user vector (0.073, 0.029, 0, 0.03) is obtained. Similar to this example, the similarity-considered user vectors of the users U02, U03, U04, and U05 for the user U02 are obtained. Here, the value of the element of the similarity-considered user vector increases as the target user has a higher degree of similarity between the user U02 and the user. For example, each element of the similarity-considered user vector of the user U05 whose inter-user similarity value is 0.98 is a value such as 0.14 to 0.98, whereas the inter-user similarity value is Each element of the similarity-considering user vector of the user U04 that is 0.029 has a value less than 0.1, and there is a significant difference in the value of the similarity-considering user vector between the user U05 and the user U04. Recognize.

  Subsequently, the content score calculation unit 12 adds, for each user, the user-considered user vectors for each user, and calculates a vector having the content score of each content as an element. A vector having the calculated content score as an element is shown in FIG. For example, the content score of the content D01 of the user U02 is the element values 0.073, 0, 0.11, 0.028 of the content D01 of the similarity-considered user vector of each user for the user U02 shown in FIG. The sum of 0.14 is calculated to be 0.35. Similarly, the content score of the content D02 of the user U02 is determined to be 0.033, the content score of the content D03 is 1.93, and the content score of the content D04 is determined to be 0.48. The content score thus obtained is a numerical value of the degree of interest and preference for each user's content, and represents the usefulness of each content for each user.

  In the present embodiment, a vector obtained by multiplying a vector representing the access frequency for each content by the similarity between users is used for calculation of the content score. Any value or vector obtained by performing an operation that reflects the similarity of the access tendency between users to a value or vector representing the access frequency of each content of the user can be used for calculation of the content score. Note that the content score calculation unit 12 may calculate the content score using the access frequency information of users whose user similarity is higher than a predetermined threshold. Even if a user vector of a user having a low similarity between users is not used for calculation of the content score, the deterioration of the accuracy of the content score is small. Therefore, by setting the predetermined threshold to an appropriate value, it is possible to reduce the processing load on the information providing apparatus 10 without deterioration of the accuracy of the content score. In this case, it is assumed that this threshold value is input to the information providing apparatus 10 in advance and stored in the content score calculation unit 12.

  The content score calculation unit 12 stores the information associated with the calculated content score and the user identification information in the index storage unit 13 as an index for use in generating the provided information (S105, content score calculation step). Specifically, the content score calculation unit 12 rearranges the elements of the content score vector for each user as shown in FIG. 11 in the descending order of the values, and further replaces them with the corresponding content IDs. The index storage unit 13 stores an index as shown in FIG. For example, it can be seen that the degree of interest and preference for the content of the user U02 decreases in the order of D03, D04, D01, and D02. In this way, by using the data in which the content IDs are arranged in the order of the content score as an index, the provision information generation process can be speeded up. In the index mode as shown in FIG. 12A, the content ID array associated with the user identification information is used as an index. The content IDs are arranged according to the size of the content score. Therefore, the content score information is substantially included. As another embodiment, as shown in FIG. 12B, a data set in which a content ID and a content score are associated can be used as an index. By using such an index, the content score can be included in the provided information when the provided information is output, so that the user presented with the provided information can quantitatively know the usefulness of each content. Furthermore, as another embodiment, a vector in which content scores are arranged as elements in ascending order of content IDs as shown in FIG. 11 can be used as an index.

  Note that the processing shown in steps S103 to S105 can be performed every time a predetermined time elapses. An arbitrary time can be set in advance as the predetermined time. These processes may be performed every time a new access history is added to the access history storage unit 22.

  Next, a processing procedure in which the information providing apparatus 10 transmits provided information to the communication terminal apparatus based on the transmitted query will be described with reference to FIG.

  First, the instruction receiving unit 31 of the communication terminal device 30 receives a query input by the user (S201). In the first embodiment, this query includes user identification information for identifying the user and a search keyword for search. The user identification information is input by the user when the user logs in to the communication terminal device 30 and stored in the user identification information storage unit 32. A plurality of search keywords may be provided, for example, or may be given by a logical expression.

  Subsequently, the instruction receiving unit 31 that has received the query transmits a query including user identification information and a search keyword to the information providing apparatus 10 via the transmission / reception unit 33 (S202).

  The query input unit 14 of the information providing device 10 receives the query transmitted from the communication terminal device 30 (S211, query input step). The query input unit 14 sends the received query to the provision information generation unit 15. Note that the query input unit 14 may acquire a query from an input device such as a keyboard or a mouse included in the information providing device 10 (for example, for maintenance purposes).

  Next, the provision information generation unit 15 refers to the index storage unit 13 and extracts an index based on the user identification information included in the received query (S212, provision information generation step). For example, when the user identification information included in the received query is the user U02, (D03, D04, D01, D02) is extracted as an index.

  Subsequently, the provision information generation unit 15 extracts content that matches the search keyword included in the query from the index extracted in step S212 (S213, provision information generation step). The search for the content including the search keyword is performed by full text search of each content included in the index. This full-text search is performed by referring to an appearance word index (not shown) for each content generated by the existing technology. This appearance word index may be stored in the index storage unit 13 or may be stored in another storage unit. Furthermore, the provision information generation part 15 produces | generates the provision information which arranged content ID of the content suitable for a search keyword in order of a content score. If the index stored in the index storage unit 13 is in the form as shown in FIG. 12A, the provision information can be generated by excluding the content ID that does not match the search keyword from the index. Further, only the content ID of the content with the highest content score among the extracted content can be used as the provided information. If the index stored in the index storage unit 13 is as shown in FIG. 11, the content ID and content score of the content that matches the search keyword are extracted, and the content IDs are arranged in descending order of the content score. , Providing information can be generated. The provision information generation unit 15 sends the generated provision information to the provision information output unit 16.

  The provided information output unit 16 transmits the acquired provided information to the communication terminal device 30 (S214, provided information output step). The transmission / reception unit 33 of the communication terminal device 30 receives the provision information and sends the received provision information to the display unit 34 (S203). The display unit 34 displays the provided provision information and presents it to the user (S204).

According to the first embodiment, the inter-user similarity calculation unit 11 of the information providing apparatus 10 calculates the inter-user similarity based on the similarity of the user's access tendency to each content. In addition, the content score calculation unit 12 extracts the access frequency for each content of each user from the access history stored in the access history storage unit 22 and weights the access frequency weighted by the similarity between users between the user and each user From this, the content score is calculated. Since the content score of the content accessed by the other user increases as the similarity between the user and the other user increases, the content score calculated in this manner accurately increases the usefulness of the document to the user. Can be represented. The provided information generation unit 15 extracts content including the search keyword included in the query, and generates provided information including the content ID of the content based on the content score. This provided information presents content that is highly useful to the user. Further, the content score is calculated from the user's access history with respect to the content, and it is not necessary to input special attribute information or the like. Therefore, it is possible to provide information useful for the user accurately and easily.
[Second Embodiment]

  FIG. 14 is a block diagram showing functional configurations of the information providing apparatus 10a, the network access server 20, and the communication terminal apparatus 30 according to the second embodiment of the present invention. An information providing apparatus 10a illustrated in FIG. 14 includes an unaccessed content information generating unit 17 instead of the provided information generating unit 15 in the information providing apparatus 10 illustrated in FIG. The other configuration of the information providing apparatus 10a is the same as that of the information providing apparatus 10.

  The unaccessed content information generation unit 17 refers to the index storage unit 13 and the access history storage unit 22 and provides only from the content identification information of the content that has not been accessed by the user specified by the user identification information included in the query. It is a means for generating information.

  A processing procedure in which the information providing apparatus 10a according to the second embodiment transmits the provided information to the communication terminal apparatus based on the transmitted query will be described with reference to FIG.

  First, the instruction receiving unit 31 of the communication terminal device 30 receives an information provision instruction input by the user (S301). This information provision instruction is performed, for example, when the user operates the search instruction button 308 of the communication terminal device 30 illustrated in FIG. Other than this operation, the user does not need to perform a special operation such as inputting a search keyword.

  The instruction receiving unit 31 that has received the information providing instruction transmits a query including only the user identification information to the information providing apparatus 10 via the transmission / reception unit 33 (S302). The user identification information is stored in the user identification information storage unit 32. For example, the user identification information may be contractor identification information stored in a SIM or UIM included in the communication terminal apparatus, or may be a cookie stored in a web browser.

  The query input unit 14 of the information providing device 10a receives the query transmitted from the communication terminal device 30 (S311, query input step). The query input unit 14 sends the received query to the unaccessed content information generation unit 17.

  Next, the unaccessed content information generation unit 17 refers to the index storage unit 13 and extracts an index based on the user identification information included in the received query (S312, provision information generation step). For example, when the user identification information included in the received query is the user U02, (D03, D04, D01, D02) is extracted as an index.

  Subsequently, the unaccessed content information generation unit 17 refers to the access history storage unit 22 and extracts the content ID of the content that has not been accessed by the user specified by the user identification information from the index extracted in step S312. (S313, provision information generation step). For example, referring to FIG. 6, the user U02 has accessed the contents D03 and D04 (the value indicating the access frequency is not zero), and has never accessed the contents D01 and D02 (shows the access frequency). Therefore, the unaccessed content information generation unit 17 extracts the content IDs (D01, D02).

  The unaccessed content information generation part 17 produces | generates what arranged extracted content ID in order of a content score as provision information (S314, provision information generation step). Further, only the content ID of the content with the highest content score among the extracted content can be used as the provided information. The unaccessed content information generation unit 17 sends the generated provision information to the provision information output unit 16.

  The provided information output unit 16 transmits the acquired provided information to the communication terminal device 30 (S315, provided information output step). The transmission / reception unit 33 of the communication terminal device 30 receives the provision information and sends the received provision information to the display unit 34 (S203). The display unit 34 displays the provided provision information and presents it to the user (S204). For example, in the communication terminal device 30 illustrated in FIG. 16, the provided information is displayed on the display 307.

According to the second embodiment, the unaccessed content information generation unit 17 extracts content that has not been accessed by the user based only on the user identification information, and further extracts content identification information and content of the extracted content. Providing information is generated based on the score. Therefore, the information providing apparatus 10 can generate and provide provided information without requesting information other than user identification information as a query. The user can obtain information on useful and unknown content. Users sometimes try to access content by performing so-called web browsing without a special purpose. In such a case, since it is difficult to come up with a search keyword as a query, the mode shown in the second embodiment is useful. In addition, the user arranges contents that have never been accessed by the user and that matches the user's interests and preferences in the order of strong compatibility by a simple operation of simply pressing the search instruction button 308 of the communication terminal device 30. Information can be obtained.
[Third Embodiment]

  FIG. 17 is a block diagram showing functional configurations of the information providing apparatus 10, the network access server 20a, and the communication terminal apparatus 30 according to the third embodiment of the present invention. The network access server 20a illustrated in FIG. 17 includes a provision information holding unit 23 in addition to the configuration of the network access server 20 illustrated in FIG. Other configurations of the network access server 20a are the same as those of the network access server 20.

  The provision information holding unit 23 is a unit that acquires and holds the provision information that the provision information output unit 16 outputs to the communication terminal device 30. The access history collection unit 21 collects an access history of access to content output from the communication terminal device 30 in response to the provided information, and is an access frequency of the content that is included in the provided information and does not collect the access history It has a function to reduce information. This access frequency information is stored in the access history storage unit 22.

  The communication terminal device 30 displays the received provision information on the display unit 34, and the user selects desired content identification information from the provision information displayed on the display unit 34, and accesses the content. This access history is collected by the access history collection unit 21 of the network access server 20a. The access history collection unit 21 refers to the provision information holding unit 23 and extracts the content identification information of the content that is included in the provision information provided to the communication terminal device 30 and is not collected as the access history. Further, the access history collection unit 21 reduces the number of access frequency information associated with the content identification information extracted here by an arbitrary number in the access history storage unit 22. This arbitrary number can be set to “1”, for example, or can be a decimal number. For example, an access history as shown in FIG. 6 is stored in the access history storage unit 22, (D03, D04) is presented as provision information to the user U 02, and the user accesses the content D 04 from the communication terminal device 30. In this case, the access frequency of the user U02 to D03 is reduced from, for example, “4” to “3”.

  According to the third embodiment, the provided information holding unit 23 holds the provided information transmitted to the communication terminal device 30, and the access history collecting unit 21 performs access history performed according to the provided information presented to the user. Therefore, it is possible to identify content that the user has not accessed despite being presented to the user as the provided information. By reducing the access frequency information stored in the access history storage unit for the content that has not been accessed in this way, the usefulness of the content for the user is more strongly reflected in the access frequency information. When the content score calculation unit 12 calculates the content score using such access frequency information, it is possible to further improve the accuracy of usability of the provided information to be provided to the user.

  Subsequently, an information providing program for causing a computer to execute a series of processes for generating an index and transmitting provided information in the first to third embodiments will be described. As shown in FIG. 18, the information providing programs 81 and 82 are stored in a program storage area 80a formed in a recording medium 80 provided in the computer.

The information providing program (for network access server) 81 includes a main module 81a that centrally controls information providing processing, an access history collecting module 81b, and a provided information holding module 81c.
The information providing program (for information providing apparatus) 82 includes a main module 82a that performs overall control of information providing processing, an inter-user similarity calculation module 82b, a content score calculation module 82c, a query input module 82d, and a provided information generation module (not yet provided). An access content information generation module) 82e and a provision information output module 82f. The functions realized by executing the access history collection module 81b and the provision information holding module 81c are the same as the functions of the access history collection unit 21 and the provision information holding unit 23, respectively. The functions realized by executing the inter-user similarity calculation module 82b, the content score calculation module 82c, the query input module 82d, the provision information generation module (unaccessed content information generation module) 82e, and the provision information output module 82f are The functions of the inter-similarity calculation unit 11, the content score calculation unit 12, the query input unit 14, the provision information generation unit 15 (or the unaccessed content information generation unit 17), and the provision information output unit 16 are the same. Note that a part or all of the information providing programs 81 and 82 may be transmitted via a transmission medium such as a communication line and received and recorded (including installation) by another device.

It is a functional block diagram of the information provision apparatus which concerns on 1st Embodiment, a network access server, and a communication terminal device. It is a hardware block diagram of an information provision apparatus. It is a hardware block diagram of a communication terminal device. It is a figure which shows the process sequence which an information provision apparatus and a network access server produce | generate an index. It is a figure which shows typically the example of the data structure of the access history stored in an access history storage part. It is a figure which shows typically the example of the data structure of the access frequency information stored in an access history storage part. It is a figure which shows typically the example of the data structure of the normalized access frequency information stored in the access history storage part. It is a figure which shows typically the example of the data structure of the user vector stored in an access history storage part. It is a figure which shows typically the example of the similarity between users which the similarity calculation part between users calculated. It is a figure which shows typically the example of the similarity consideration consideration user vector which the similarity calculation part between users calculated. It is a figure which shows typically the example of the content score which the content score calculation part computed. It is a figure which shows typically the example of the data structure of the index stored in an index storage part. It is a figure which shows the process sequence which the information provision apparatus which concerns on 1st Embodiment transmits provision information to a communication terminal device based on the transmitted query. It is a functional block diagram of the information provision apparatus which concerns on 2nd Embodiment, a network access server, and a communication terminal device. It is a figure which shows the process sequence which the information provision apparatus which concerns on 2nd Embodiment transmits provision information to a communication terminal device based on the transmitted query. It is a figure which shows the example of a communication terminal device typically. It is a functional block diagram of the information provision apparatus which concerns on 3rd Embodiment, a network access server, and a communication terminal device. It is a figure which shows the structure of the information provision program which concerns on embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10, 10a ... Information provision apparatus, 11 ... Inter-user similarity calculation part, 12 ... Content score calculation part, 13 ... Index storage part, 14 ... Query input part, 15 ... Provision information generation part, 16 ... Provision information output part, 17 ... Unaccessed content information generation unit, 20, 20a ... Network access server, 21 ... Access history collection unit, 22 ... Access history storage unit, 23 ... Provided information holding unit, 30 ... Communication terminal device, 31 ... Instruction reception unit, 32 ... User identification information storage unit, 33 ... Transmission / reception unit, 34 ... Display unit, 40 ... Network, 81, 82 ... Information provision program, 81b ... Access history collection module, 81c ... Provision information holding module, 82a ... Main module, 82b ... Inter-user similarity calculation module, 82c ... Content score calculation module, 82d ... Query input Module, 82e ... provides information generation module (not access the content information generating module), 82f ... provides information output module.

Claims (10)

Access history storage means for storing access frequency information representing the frequency of access to each content by a user in association with user identification information for identifying the user;
An inter-user similarity calculating unit that calculates an inter-user similarity representing the similarity of the access tendency to the content between the users based on the access frequency information stored in the access history storage unit;
Content score calculation means for calculating a content score, which is information representing the usefulness of content for the user, from the access frequency information of each user weighted by the similarity between users between the user and each user;
Index storage means for storing the content score of each content calculated by the content score calculation means in association with the user identification information;
Query input means for receiving an input of a query including user identification information transmitted from a communication terminal device;
The content identification information of the content that matches the query accepted by the query input means is acquired, and the content score stored in the index storage means in association with the user identification information included in the query is referred to Providing information generating means for generating providing information from the acquired content identification information;
An information provision apparatus comprising: provision information output means for outputting the provision information generated by the provision information generation means to the communication terminal apparatus. The provision information generation means provides only the content identification information of the content that has not been accessed by the user specified by the user identification information included in the query with reference to the index storage means and the access history storage means The information providing apparatus according to claim 1, wherein information is generated. It further comprises access history collection means for collecting a history of access to each content by a user, generating access frequency information based on the collected access history, and storing the generated access frequency information in the access history storage means. The information providing apparatus according to claim 1 or 2. The access history collection means collects an access history of access to content from the communication terminal apparatus in response to the provision information output to the communication terminal apparatus by the provision information output means, and is included in the provision information 4. The information providing apparatus according to claim 3, wherein access frequency information of content not included in the access history is reduced. The information providing apparatus according to claim 1, wherein the provision information generating unit generates provision information in which the content identification information is arranged in order of the content score. The query transmitted from the communication terminal device includes a search keyword,
The information providing apparatus according to claim 1, wherein the provided information generating unit acquires content identification information of content including the search keyword. The information providing apparatus according to claim 1, wherein the content identification information includes information indicating a location of the content on a network. The content score calculation means calculates the content score using the access frequency information of the user whose similarity between the users is higher than a predetermined threshold. Information providing device In the information providing method in the information providing apparatus,
The information providing device is
Content between users based on the access frequency information stored in the access history storage means for storing access frequency information indicating the frequency of access to each content by the user in association with user identification information for identifying the user An inter-user similarity calculating step for calculating an inter-user similarity indicating the similarity of access tendencies to
A content score, which is information representing the usefulness of content for the user, is calculated from the access frequency information of each user weighted by the similarity between the users between the user and each user. A content score calculation step of storing the content score in the index storage means in association with the user identification information;
A query input step for receiving an input of a query including at least user identification information transmitted from the communication terminal device;
A provision information generation step of acquiring content identification information of the content conforming to the query received by the query input step, and generating provision information from the acquired content identification information with reference to the index storage means;
A provision information output step of outputting the provision information generated in the provision information generation step to the communication terminal device; Computer
Access history storage means for storing access frequency information representing the frequency of access to each content by a user in association with user identification information for identifying the user;
An inter-user similarity calculating unit that calculates an inter-user similarity representing the similarity of the access tendency to the content between the users based on the access frequency information stored in the access history storage unit;
Content score calculation means for calculating a content score, which is information representing the usefulness of content for the user, from the access frequency information of each user weighted by the similarity between users between the user and each user;
Index storage means for storing the content score of each content calculated by the content score calculation means in association with the user identification information;
Query input means for receiving an input of a query including at least user identification information transmitted from the communication terminal device;
Providing information generating means for acquiring content identification information of the content conforming to the query received by the query input means, referring to the index storage means, and generating providing information from the acquired content identification information;
An information provision program for causing the provision information generated by the provision information generation means to function as provision information output means for outputting to the communication terminal device.

Images