JP5598265B2 - Information providing apparatus, method and program - Google Patents

Description

  The present invention relates to an information providing apparatus, an information providing method, and an information providing program for providing information to a user.

  In recent years, attempts have been made to increase communication opportunities among people who have gathered by chance in halls at public baths and hospital conversation spaces, and to create and deepen connections between people.

  As a technology related to this, for example, in Patent Document 1, information on which a user is likely to show interest is extracted from the information browsing history of each user based on the number of times the same URL is viewed, the appearance frequency of words, and the like. The technology presented to the user is described.

  Non-Patent Document 1 describes a technique for creating a user group and an information cluster by collecting users and information using similarities between features of interest of users or features of information. .

JP 2008-97078 A

Toby Segaran, “Collective Intelligence Programming”, O'Reilly Japan, published on 25th July 2008, p. 31-58

  For example, an apparatus has been proposed in which some information is provided to people gathered in a certain place to give a conversation opportunity and activate people's conversation. In order to realize such an apparatus, for example, it is conceivable to use the following method by combining the technique described in Non-Patent Document 1 and the technique described in Patent Document 1.

・ Method 1
(1) Assuming that each user's interests are reflected in the browsing information, the use of similar interests from information browsed by users in the past using the method described in Non-Patent Document 1. Group users together to create a user group.

  (2) Using the method described in Patent Document 1, information including words and the like common to information viewed by many users in the user group is provided to the user group.

  By using this method, it is possible to provide information that many users in the user group are likely to be interested in.

  In the above method, a group of users who see similar information is created, and information is provided according to the statistical interest of the users in each user group. As another method, for example, information having similar contents can be gathered as a cluster, and people who are interested in the information included in each information cluster can be gathered to provide the information.

・ Method 2
(1) Using the method described in Non-Patent Document 1, information is clustered from information similarity to create an information cluster.

(2) From the similarity between the information included in the information cluster and the feature of interest of the user, users who prefer the information cluster are extracted and a user group is created.

(3) Providing information included in the information cluster to the user group.

  As described above, the above-described apparatus can be realized by applying the method 1 or the method 2.

  However, even if information is provided to a user registered in the apparatus, the interest of the user varies from user to user. Further, the number of information that can be collected within a certain time is limited.

  Therefore, even if one or more user groups are created by looking at the similarities between user interests, information that matches the interests common to all users in each user group may not be provided. There may be users in the user group who are not interested in the information provided.

  Also, looking at the similarity between information, creating one or more information clusters does not necessarily create an information cluster composed of information that meets the interests common to all users belonging to the user group. There may be users in the user group who are not interested in the provided information.

  For example, when the method 1 is used, there is a possibility that the user group is not properly divided. For example, consider the case where there are users who are interested in the following.

・ User A: Interested in soccer and Japanese politics.
-User B: Not only soccer but also interested in sports such as baseball and golf, and is also interested in Japanese music.
・ User C: Interested in soccer but only in the J-League. Also interested in music.
・ User D: Interested in baseball and international politics.
User E: Interested in baseball, but only interested in Japanese professional baseball. Also interested in Japanese music.

  It is assumed that the user group (A, B, C) and the user group (D, E) are created using the method 1. Based on the interests of each user belonging to the user group (A, B, C), the information of the J League considered that all users in the user group may be interested in the user group (A, B, C) ) Is considered to activate user conversation.

  Moreover, it is thought that a user's conversation will be activated if a Japanese professional baseball information can be provided with respect to a user group (D, E).

  However, if there is information on Japan-Korea relations and Japanese music in the information list, but there is no information on J-League or Japanese professional baseball, information that everyone in the user group is interested in Can not provide. In such a case, those who divide users into user groups (A, D) and user groups (B, C, E) and provide information on Japanese international politics and information on Japanese music, respectively. However, it is thought that conversation is activated.

  The same thing can happen when information is clustered first. For example, consider the following five pieces of information.

・ Information 1: Pro golfer A, new hairstyle “Asymmetry” is displayed ・ Information 2: B player returns to Japan in “season-first” fashion ・ Information 3: Italian brand C collaborates with designer D ・ Information 4: “Worst. Isn't it a company product? President of Italian luxury brand reports and information 5: Resignation from Ministry of Education, Ministry of Education, etc.

  Assume that method 2 is used to create an information cluster (1, 2, 3) about fashion and an information cluster (4, 5) about politics. In this case, if fashion information can be provided to a user group interested in fashion and political information can be provided to a user group interested in politics, it is considered that user conversation is activated.

  However, there are users who are interested in sports and users who are interested in Italy, but there are no users who are interested in fashion and politics. , Users who belong to user groups cannot provide information that interests them.

  In such a case, it is considered that the user's conversation is activated when the information cluster is the sports information cluster (1, 2, 5) and the Italian information cluster (3, 4).

  As described above, even if a user group or an information cluster is created based only on the interests or information of the users of the above-mentioned devices in the same place, information that interests all users is provided. It is not possible.

  Therefore, the present invention groups users so that there are no people who are not interested in the information provided when providing information that is a topic of conversation to people gathered in the same place, It is an object of the present invention to provide an information providing apparatus, an information providing method, and an information providing program for providing information that all users belonging to a user group are interested in.

An information providing apparatus according to the present invention receives an information vector representing a feature of information represented by a character string and a user interest vector representing a feature of information viewed by the user, and classifies the user, An information providing device that creates a user group and provides information to the user group, estimating the degree of interest for each user for each information from the similarity between the information vector and the user interest vector, For information, the information interest vector creating means for creating an information interest vector having a user as a dimension and the degree of interest as a value, an information cluster creating means for clustering information using the information interest vector, and an information cluster in contrast, the fact that a higher degree user interests, and a user extracting means for extracting from the information vector in the information cluster to create a user group And butterflies.

An information providing apparatus according to the present invention receives an information vector representing a feature of information represented by a character string and a user interest vector representing a feature of information viewed by the user, and classifies the user, An information providing device that creates a user group and provides information to the user group, estimating the degree of interest for each user for each information from the similarity between the information vector and the user interest vector, Information interest vector creation means for creating an information interest vector having a user as a dimension and the degree of interest as information, information cluster creation means for clustering information using the information interest vector, and created information presents data representing features of a cluster to the user, according to user's selection operation, characterized in that a user selection means for selecting information cluster corresponding to the user To.

According to the information providing method of the present invention, a computer inputs an information vector representing a feature of information represented by a character string and a user interest vector representing a feature of information viewed by the user, and classifies the user. An information providing method for creating a user group and providing information to the user group, wherein the computer determines the degree of interest of each user for each piece of information by comparing the information vector with the user interest vector For each piece of information, create an information interest vector with each user as the dimension and the degree of interest as the value, cluster the information using the information interest vector, A user having a high degree is extracted from an information vector in the information cluster in order to create a user group . In the information providing method according to the present invention, the computer inputs an information vector representing the characteristics of the information represented by the character string and a user interest vector representing the characteristics of the information browsed by the user. Is an information providing method for creating a user group and providing information to the user group, wherein the computer determines the degree of interest of each user for each information as an information vector and a user interest vector. For each piece of information, create an information interest vector with the user as the dimension and the degree of interest as the value, cluster the information using the information interest vector, and the characteristics of the created information cluster Is displayed to the user, and an information cluster corresponding to the user is selected according to the user's selection operation.

An information providing program according to the present invention inputs, to a computer, an information vector representing a feature of information represented by a character string and a user interest vector representing a feature of information viewed by the user, and classifies the user. An information providing program for creating a user group and executing a process for providing the user group with information, and using the information vector and the degree of interest of each user for each information Information interest vector creation processing that creates an information interest vector with a user as a dimension and a degree of interest as a value, and information using the information interest vector. Information cluster creation processing for clustering users and users who are highly interested in the information cluster in order to create user groups. Characterized in that to execute the user extraction process of extracting the information vector in register. In addition, the information providing program according to the present invention inputs an information vector representing the characteristics of the information represented by a character string and a user interest vector representing the characteristics of the information viewed by the user to the computer. Is an information providing program for executing a process of creating a user group and providing information to the user group, wherein the information vector indicates the degree of interest of each user for each piece of information. Information interest vector creation processing that creates an information interest vector that takes the user as a dimension and the degree of interest as a value for each information, and the information interest vector The information cluster creation process for clustering information and the data representing the characteristics of the created information cluster are presented to the user, and the user is selected according to the user's selection operation. Characterized in that to execute a user selecting process of selecting the corresponding information cluster's.

  According to the present invention, when providing information that is a topic of conversation to people gathered in the same place, users are grouped so that there is no person who is not interested in the information to be provided. Can provide information that all users who belong to a group of users are interested in.

It is a block diagram which shows an example of a structure of 1st Embodiment of an information provision apparatus. It is explanatory drawing which shows an example of the data which an information word frequency memory | storage part stores. It is explanatory drawing which shows an example of the data which a browsing word frequency memory | storage part stores. It is a flowchart which shows the example of operation | movement of 1st Embodiment of an information provision apparatus. It is a block diagram which shows an example of a structure of 2nd Embodiment of an information provision apparatus. It is a flowchart which shows the example of operation | movement of 2nd Embodiment of an information provision apparatus. It is a block diagram which shows the minimum structural example of an information provision apparatus.

Embodiment 1. FIG.
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a configuration example of the first embodiment of the information providing apparatus. In the present embodiment, data that can be expressed as text is defined as information.

  In the present embodiment, the information providing apparatus includes an information storage unit 100, a user interest storage unit 101, a user group creation unit 102, and an information output unit 106. The information providing apparatus is specifically realized by an information processing apparatus such as a personal computer that operates according to a program.

  The information storage unit 100 stores information and information features. Hereinafter, the expression “store the feature” is used. Specifically, the data indicating the feature is stored. Specifically, the information storage unit 100 is realized by a storage device such as a magnetic disk device or an optical disk device. FIG. 2 shows an example of data stored in the information storage unit 100.

  In the example shown in FIG. 2, morphological analysis is performed on information in advance to determine the appearance frequency of each word (noun in the example of FIG. 2). The information storage unit 100 stores information features in the form of vectors (herein referred to as information vectors) having words as dimensions and appearance frequencies as values. These pieces of information are registered in advance in the information storage unit 100 by, for example, an administrator. In addition, for example, the information may be collected periodically from a server operated by the media, an advertisement distribution company, or the like, analyzed by morpheme, and accumulated.

  The user interest storage unit 101 stores characteristics of information that the user has shown interest in. Specifically, the user interest storage unit 101 is realized by a storage device such as a magnetic disk device or an optical disk device. FIG. 3 shows an example of data stored in the user interest storage unit 101.

  In the example shown in FIG. 3, it is assumed that the user browses information of interest, morphological analysis is performed on the information browsed by a certain user, and the appearance frequency of each word (in the example of FIG. 3) is obtained. ing. The user interest storage unit 101 stores the features of interest of each user in the form of a vector having a word as a dimension and an appearance frequency as a value (herein referred to as an interest vector). Such information is registered in the user interest storage unit 101 in advance by, for example, an administrator. Further, for example, a user's search history may be periodically collected using a system such as a provider, and morphological analysis may be performed to accumulate these pieces of information.

  The user group creation unit 102 creates a list of information (information cluster) by estimating the degree of interest of each user for each information, and creates user group data by extracting user data from each information cluster. It has a function to do.

  Hereinafter, data specifying an actual user is referred to as user data, and data specifying a user group is referred to as user group data. The user group is not a group of users that actually exist at the place where the information providing apparatus is installed, but a set of users determined by the information providing apparatus as a target for providing certain information. In this embodiment, the expression that a user group is created is used. Specifically, user group data for specifying a user group is created.

  Specifically, the user group creation unit 102 is realized by a CPU of an information processing apparatus that operates according to a program. The user group creation unit 102 includes an information interest vector creation unit 103, an information cluster creation unit 104, and a user extraction unit 105.

  Here, the degree of interest in the user's information is called the interest level. The information interest vector creation unit 103 estimates the interest level of each user for each piece of information from the similarity between the information vector and the interest vector, and for each piece of information, a vector having the user as a dimension and the interest level as a value. (Hereinafter also referred to as an information interest vector). Note that the information interest vector creation unit 103 can obtain the degree of interest using an existing method, for example, by measuring the similarity between the information vector and the interest vector as a similarity measurement method.

  The information cluster creation unit 104 has a function of clustering information interest vectors and collecting information indicating similar interests of users to create N clusters (referred to herein as information clusters). Note that the information cluster creation unit 104 can use an existing method such as hierarchical clustering or k-means clustering as a clustering method. Further, for example, the number N of information clusters to be created may be determined based on the number of display devices such as a display device used when providing information.

  The user extraction unit 105 forms a user group by assigning the information included in each information cluster to a user group from a highly interested user, and creates a pair of the user group and the information cluster. It has a function.

  For example, (1) the user extraction unit 105 extracts user data in descending order of interest with respect to information included in the information cluster, and generates user group data including the extracted user data. Then, the user extracting unit 105 associates the generated user group data with the extraction source information cluster.

  Further, for example, (2) the user extraction unit 105 extracts user data in descending order of the degree of interest in the information included in the information cluster, and associates the extracted user data with the extraction source information cluster in advance. Correlate with user group data.

  Further, for example, (3) the user extraction unit 105 extracts user data in descending order of interest with respect to information included in the information cluster, and associates the extracted user data with the extraction source information cluster. The set of the obtained user data is defined as user group data for the information cluster.

  Hereinafter, expressions such as “assign user data to user group data” and “assign user data to user group data” are used. Specifically, the user extraction unit 105 performs the above (1). ) To (3).

  The information output unit 106 has a function of providing information included in the corresponding information cluster to the user group. The information output unit 106 provides information to a user group by controlling the information output unit 106 to display the information on a display device such as a display device, for example. Further, for example, the information output unit 106 distributes information to terminal devices held by users included in the user group.

  In the present embodiment, an example will be described in which a plurality of users exist in a place where the information providing apparatus is installed, and information is provided to the plurality of users by displaying the information on the display device. Note that the present invention is not limited to such a form, and for example, the information providing apparatus may be a form in which information is distributed to terminal devices held by users. In this case, for example, it is assumed that the user interest storage unit 101 accumulates information indicating contents viewed by a user on a website or the like using a terminal device and stores the information in the form of an interest vector.

  Next, the operation of the information providing apparatus will be described with reference to FIG. FIG. 4 is a flowchart illustrating an example of the operation of the information providing apparatus according to the first embodiment. For example, the information providing apparatus executes the following operation in accordance with the operation of the user or administrator.

  For example, when the administrator performs an activation operation of the information providing apparatus, the information interest vector creation unit 103 extracts an information vector of information from the information word frequency storage unit 100 (step S1).

  Next, the information interest vector creation unit 103 extracts the user's interest vector from the user interest storage unit 101 (step S2).

  Next, the information interest vector creation unit 103 estimates the degree of interest from the extracted information vector and interest vector (step S3). That is, the information interest vector creation unit 103 calculates the degree of interest based on the extracted information vector and interest vector.

  Specifically, in the example illustrated in FIGS. 2 and 3, the information interest vector creation unit 103 can use the similarity of the appearance frequency of words between the information vector and the interest vector as the degree of interest. For details of specific methods such as cosine similarity for similarity calculation, see, for example, literature (Takenobu Tokunaga, “Information Retrieval and Language Processing (Language and Calculation)”, University of Tokyo Press, November 1999, p. .31).

  Thereafter, the information interest vector creation unit 103 repeats the processing from step S1 to step S3 until the degree of interest has been estimated for all combinations of information vectors and interest vectors (step S4).

  Next, the information interest vector creation unit 103 creates an information interest vector with the user as the dimension and the interest level as the value for the information (step S5).

For example, if the interest levels of users A, B, C, D, and E are 0.2, 0.8, 0.1, 0.5, and 0.3 for information 1, respectively, The information interest vector v ₁ created by the vector creation unit 103 is v ₁ = (0.2, 0.8, 0.1, 0.5, 0.3).

  Next, the information cluster creation unit 104 clusters the information interest vectors and collects the information into N information clusters according to the user's interest (step S6). In other words, the information cluster creation unit 104 clusters information interest vectors, collects information indicating similar interests of users, and creates N information clusters.

  Next, the user extraction unit 105 extracts user data indicating users with a high degree of interest in the information included in the information cluster, and all user data is always user group data corresponding to the information cluster. Distribution is made so as to belong to any one of them, and user group data is created (step S7).

  For example, the user extraction unit 105 obtains and standardizes the sum of information interest vectors in each information cluster, calculates an average interest vector, and uses each value of the average interest vector (average interest of each user) high. The user group data corresponding to the information cluster is assigned in order from the user data. The user extracting unit 105 can create user group data by repeating this process until all user data belongs to some user group data. At this time, the user extraction unit 105 performs control so that user data already assigned to some user group data is not assigned to other user group data, and one user data is one use. To belong only to user group data.

  By executing the above processing, the user group, which is a set of users specified based on the degree of interest, can be associated with the information cluster. Therefore, the information output unit 106 supports the user group. The information included in the information cluster to be provided is provided (step S8). Specifically, the information output unit 106 provides information to the user group by performing control so that information is displayed on a display device such as a display device.

  As described above, in the present embodiment, information, information characteristics, and user interest characteristics are input, information is clustered based on each user's degree of interest in the information, User group data is created by assigning users in descending order of interest. At that time, user group data is created so that all user data must belong to one of the user group data. Then, the information included in the corresponding information cluster is provided to the user group.

  Accordingly, it is possible to realize an information providing apparatus that groups users so that no one is not interested in the information to be provided and provides information that all users belonging to each user group are interested in be able to.

Embodiment 2. FIG.
Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a block diagram illustrating a configuration example of the second embodiment of the information providing apparatus.

  As shown in FIG. 5, this embodiment is different from the first embodiment in that a user selection unit 107 is included instead of the user extraction unit 105. Other components are the same as those in the first embodiment. Only differences from the first embodiment will be described below.

  The user selection unit 107 extracts user data from the information cluster in descending order of interest, and assigns the extracted user data to user group data. All user data is assigned to any user group data. Has the ability to repeat until you belong.

  Here, unlike the user extraction unit 105 of the first embodiment, the user selection unit 107 can make user data belong to a plurality of user group data.

  Specifically, when the user selection unit 107 assigns user data to a plurality of information clusters (that is, user group data) in the process of repeating the assignment, the user group data once assigned as it is. To the user data.

  In addition, after the assignment to the user group is completed, the user selection unit 107 presents meta information such as feature words representing information included in the information cluster to users belonging to a plurality of user groups. The user can select which user group the user belongs to.

  Specifically, the user selection unit 107 performs control so that meta information of a plurality of information clusters is displayed on a display device such as a display device. When the user who has seen the meta information performs an operation of selecting one of the information clusters (that is, the user group to which the user belongs), the user selection unit 107 performs corresponding user data according to the user's operation. Allocate only to the user group data selected.

  Next, the operation of this embodiment will be described. FIG. 6 is a flowchart illustrating an example of an operation of the information providing apparatus according to the second exemplary embodiment.

  The example of the operation of this embodiment shown in FIG. 6 is obtained by adding steps S9 and S10 to the example of the operation of the first embodiment shown in FIG. Since the processing of steps S1 to S6 is the same as that of the first embodiment, only the processing after step S7 will be described.

  When an information cluster is created in step S6, the user selection unit 107 extracts user data from the information cluster in descending order of interest, and assigns the extracted user data to user group data (step S7). Here, unlike the first embodiment, the user selection unit 107 does not control one user data so that it belongs only to one user group data, and the user data is stored in a plurality of user groups. Can belong to data.

  Next, the user selection unit 107 determines whether there are users belonging to a plurality of user groups (step S9). That is, the user selection unit 107 determines whether user data included in a plurality of user group data exists.

  If it is determined that there are users belonging to a plurality of user groups, the user selection unit 107 gives meta information about the information cluster corresponding to the user groups to which the user belongs. Present (step S10). Specifically, the user selection unit 107 performs control so that meta information of a plurality of information clusters is displayed on a display device such as a display device.

  The user selection unit 107 extracts, for example, feature words of M words from each information cluster by using an index such as TF-IDF in order to present meta information representing information included in the information cluster to the user. Present characteristic words to the person. By executing such processing, the user can know what information is included in each information cluster. Note that TF-IDF for extracting feature words is described in, for example, a document (Takenobu Tokunaga, “Information Retrieval and Language Processing (Language and Calculation)”, University of Tokyo Press, November 1999, p. 32). Have been described.

  As a result, users belonging to multiple user groups can know what information is included in the information cluster, so select the information according to the user's preference at that time, Information can be provided.

  When the user who has seen the meta information performs an operation of selecting one of the information clusters (that is, the user group to which the user belongs), the user selection unit 107 performs corresponding user data according to the user's operation. Allocate only to the user group data selected.

  Next, the information output unit 106 provides the information included in the corresponding information cluster to the user group based on the reassigned user group data (step S8). Specifically, the information output unit 106 provides information to the user group by performing control so that information is displayed on a display device such as a display device.

  As described above, in this embodiment, in addition to the effects of the first embodiment, by presenting and selecting a plurality of information clusters having a high degree of interest for the user, information that more matches the user's interest Can be provided.

  As described above, according to the present invention, information is clustered based on the degree of interest of each user with respect to information, and a user group is created by assigning users to the information cluster in descending order of interest. To do. At that time, a user group is created so that all users are always assigned to one of the information clusters. Therefore, users can be grouped so that there is no person who is not interested in the information to be provided, and information that all users belonging to individual user groups can be interested in can be provided.

  Next, the minimum configuration of the information providing apparatus according to the present invention will be described. FIG. 7 is a block diagram illustrating a minimum configuration example of the information providing apparatus. As shown in FIG. 7, an information providing apparatus that provides information to a user group into which users are classified includes a user group creating means 1 as a minimum component.

  In the information providing apparatus with the minimum configuration shown in FIG. 7, the user group creation means 1 estimates the degree of interest in the user's information, classifies the information based on the degree of interest, creates an information cluster, Create user groups based on clusters and degree of interest.

  Therefore, according to the information providing apparatus with the minimum configuration, when providing information that becomes a topic of conversation to users gathered in the same place, there is no user who is not interested in the information to be provided. Can be grouped to provide information that all users belonging to individual user groups are interested in.

  In this embodiment, the characteristic configuration of the information providing apparatus as shown in the following (1) to (5) is shown.

  (1) An information providing apparatus is an information providing apparatus that provides information to a user group into which users are classified, and estimates the degree of interest (for example, degree of interest) of the user's information. A user group creation unit (for example, realized by the user group creation unit 102) that creates an information cluster by classifying information based on the information cluster and a degree of interest is provided. It is characterized by that.

  (2) The information providing apparatus inputs information represented by a character string, a feature element set of information, and user interest information representing the user's interest in the information feature element set, and classifies the user. An information providing apparatus that creates a user group and provides information to the user group, and the degree of interest in each piece of information for each user based on the information, the feature element set, and the user interest information ( For example, a user group creation means (for example, use) that estimates an interest level, classifies information according to the degree of interest, creates an information cluster, and creates a user group based on the degree of interest from the information cluster A person group creation unit 102).

  (3) In the information providing device, the feature element is an information vector having a word appearing in the information as a dimension and the word appearance frequency as a value, and the user interest information is a word in the information viewed by the user Is a user interest vector whose value is the appearance frequency in the information viewed by the user of the word, and the user group creation means uses each information based on the information vector and the user interest vector. Information interest that estimates the degree of interest for each person from the similarity between the information vector and the user interest vector, and creates an information interest vector for each piece of information with the user as the dimension and the degree of interest as the value Vector creation means (for example, realized by the information interest vector creation unit 103) and information cluster creation means for clustering information using the information interest vector (for example, information cluster) And a user extraction means (for example, a user extraction means for extracting a user who is highly interested in an information cluster from an information vector in the information cluster in order to create a user group). (Implemented by the user extraction unit 105).

  (4) In the information providing apparatus, the user group creating means is configured such that a word appearing in information used as a feature element is a dimension, an information vector having a word appearance frequency as a value, and a user used as user interest information. Based on the user interest vector having the word in the browsed information as a dimension and the appearance frequency in the information browsed by the user of the word as a value, the degree of interest for each user for each information is expressed as an information vector. Information interest vector creation means (for example, an information interest vector creation unit) that creates an information interest vector that is estimated from the similarity to the user interest vector and takes the user as a dimension and the degree of interest as a value for each information 103) and information cluster creation means for clustering information using the information interest vector (for example, the information cluster creation unit 104 The user selection means (for example, the user selection unit 107) presents data representing the characteristics of the created information cluster to the user and selects an information cluster corresponding to the user according to the user's selection operation. May be included).

  (5) The information providing apparatus includes display control means (for example, realized by the information output unit 106) that controls the display apparatus to display information included in the information cluster corresponding to the created user group. It may be configured.

  INDUSTRIAL APPLICABILITY The present invention can be applied to an application that provides an environment that makes it easy to talk with people around and knows strangers and makes friends better.

DESCRIPTION OF SYMBOLS 1 User group creation means 100 Information word frequency memory | storage part 101 User interest memory | storage part 102 User group creation part 103 Information interest vector creation part 104 Information cluster creation part 105 User extraction part 106 Information output part 107 User selection part

Claims (8)

An information vector representing the characteristics of the information represented by the character string and a user interest vector representing the characteristics of the information browsed by the user are input, classify the users, create a user group, An information providing device for providing information to a user group,
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. An information interest vector creating means for creating an information interest vector,
Information cluster creating means for clustering the information using the information interest vector;
User providing means for extracting users who are highly interested in the information cluster from information vectors in the information cluster in order to create a user group apparatus. An information vector representing the characteristics of the information represented by the character string and a user interest vector representing the characteristics of the information browsed by the user are input, classify the users, create a user group, An information providing device for providing information to a user group,
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. An information interest vector creating means for creating an information interest vector,
Information cluster creating means for clustering the information using the information interest vector;
It comprises user selection means for presenting data representing the characteristics of the created information cluster to the user and selecting an information cluster corresponding to the user in accordance with the selection operation of the user. Information providing device. The information vector is a vector having a word appearing in the information as a dimension and the appearance frequency of the word as a value,
The user interest vector is a vector whose dimension is a word in the information viewed by the user and whose value is the appearance frequency of the word in the information browsed by the user.
The information providing apparatus according to claim 1 or 2. Provided with a display control means for controlling the display device to display information included in the information cluster corresponding to the created user group
The information providing apparatus according to any one of claims 1 to 3 . The computer inputs an information vector representing the characteristics of the information represented by a character string and a user interest vector representing the characteristics of the information viewed by the user, classifies the users, and sets a user group. An information providing method for creating and providing information to a user group,
The computer is
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. Create an information interest vector
Clustering the information using the information interest vector;
A method for providing information , wherein a user who is highly interested in the information cluster is extracted from an information vector in the information cluster in order to create a user group . The computer inputs an information vector representing the characteristics of the information represented by a character string and a user interest vector representing the characteristics of the information viewed by the user, classifies the users, and sets a user group. An information providing method for creating and providing information to a user group,
The computer is
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. Create an information interest vector
Clustering the information using the information interest vector;
The data representing the characteristics of the created information cluster is presented to the user, and the information cluster corresponding to the user is selected according to the user's selection operation.
An information providing method characterized by the above. An information vector representing the characteristics of the information represented by the character string and a user interest vector representing the characteristics of the information browsed by the user are input to the computer, the users are classified, and a user group is defined. An information providing program for creating and executing a process for providing information to a user group,
In the computer,
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. An information interest vector creation process for creating an information interest vector,
An information cluster creation process for clustering the information using the information interest vector;
An information providing program for executing, on the information cluster, a user extraction process of extracting a user having a high degree of interest from an information vector in the information cluster in order to create a user group . An information vector representing the characteristics of the information represented by the character string and a user interest vector representing the characteristics of the information browsed by the user are input to the computer, the users are classified, and a user group is defined. An information providing program for creating and executing a process for providing information to a user group,
In the computer,
The degree of interest for each user with respect to each piece of information is estimated from the similarity between the information vector and the user interest vector, and for each piece of information, the user is taken as a dimension, and the degree of interest is a value. An information interest vector creation process for creating an information interest vector,
An information cluster creation process for clustering the information using the information interest vector;
User selection processing for presenting data representing the characteristics of the created information cluster to the user, and selecting an information cluster corresponding to the user according to the user's selection operation.
An information provision program for execution.

Images