JP2003099458A - Contents providing system, contents providing server, its program and recording medium with the program recorded thereon - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a contents providing system, a contents providing server, its program and a recording medium with the program recorded thereon to enable an agent to collect information and to provide contents suitable for tastes and needs of a user. SOLUTION: In the contents providing system in which the user, a contents provider who provides the user with the contents and the contents providing server that provides the user with contents data regarding the contents of the contents provider are connected with one another by a communication network, the above problem is solved by providing it with a data base that stores the information to select the contents data suitable for the user, the agent that selects the contents data suitable for the user by using the data base and an agent that acquires the selected contents data suitable for the user from the contents provider.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a content providing system, a content providing server, a program therefor, and a recording medium having the program recorded therein, and more particularly to a content providing system for providing content suitable for a user, and a content providing system. The present invention relates to a server, a program thereof, and a recording medium recording the program.

[0002]

2. Description of the Related Art In recent years, with the development of information infrastructure, more multimedia information has been distributed to viewers (users) by digital satellite broadcasting, the Internet and the like. As a result, the number and types of programs (contents) that can be used by the viewers are rapidly expanding, and it is difficult for the viewers to properly select a desired program from a wide variety of programs.

In order to solve the above problem, there is a conventional program search system (search engine) using a user profile as shown in FIG.

FIG. 1 is a diagram of a first configuration example of a conventional program search system.

The program search system shown in FIG. 1 is configured to include an interface 10, an interface processing section 11, a system management operation section 12, an inference engine 13, and a user profile processing section 14.

Here, in general, a user profile is
The personal basic information of the user, such as name, age, gender, occupation, address, nationality, telephone number, e-mail address, hobby, interest, language used, connection means (BS (Broadcast
ing Satellite), cable, etc.), payment method, favorite program genre, etc. Further, in the present invention, history information of programs actually watched by each user is also stored as a user profile. The history information of a program includes items such as the program genre viewed, the viewing time zone, the date, and the day of the week.

The user profile processing unit 14 and the inference engine 13 shown in FIG. 1 include, for example, pre-event variables such as "viewing day of the week" and "viewing time zone" existing in the user profile, and a reference variable of "program genre". It has an inference function to derive each other. That is, in order to correct the reference variable while inferring the reference variable from the prior event variable by providing a knowledge expression representing a causal relationship peculiar to a given region,
It has the function of tracing back the reference variable from the prior event variable.

[0008] First, the user selects the user interface 1
When personal basic information is registered in 0, the user profile processing unit 14 stores the user profile via the network. The user can change the registered content,
The changed data is reflected in the database of the user profile processing unit 14.

Next, when the user makes an inquiry to the program search system about a related program that suits his / her hobby, the interface processing section 11 recognizes it, and the inference engine 13
Send instructions to. The inference engine 13 has a program information database 15, and from the user profile processing unit 14,
User profile information of the instructed user is extracted, related program information is selected from the extracted information, and output to the interface processing unit 11. The interface processing unit 11 edits the display content so that the user can easily recognize it, and displays the edited content on the interface 10. The user views the desired program from the displayed program information list. The programs actually viewed by the user are stored in the database as viewing history information from the user profile processing unit 14. The system management operation unit 12 controls a series of flows.

Here, the program information database 15 is
The program net time, the program name, the actual program, etc. of various program genres are stored in the inference engine in an amount that can be stored.

By constructing the program search system as described above, a database specialized for the user can be constructed and program information relevant to the user can be provided.

However, in the system as shown in FIG. 1, the program is provided mainly within the range of the user profile. Therefore, even if there is a program that the user potentially likes, it will be difficult to pick it up anew. In addition, the combination of selected programs lacks novelty and tends to be uniform. Therefore, the user becomes tired because the service becomes less attractive and less attractive.

Therefore, in order to newly pick up a program that is potentially suitable for tastes and needs, a user profile for managing data specialized for personal information and a public opinion survey are provided in the part for managing personal data. It is conceivable that a so-called user model in which some models are created by using statistical data indicating viewing habits such as, and the user model exists as interpolation data of the user profile.

A program search system including the above contents is shown in FIG.

FIG. 2 is a diagram of a second configuration example of a conventional program search system.

The program search system of FIG. 2 is provided with an interface 10, an interface processing unit 11, a system management operation unit 12, an inference engine 13, a user profile processing unit 14, and a user model processing unit 20. It is configured.

In the user model, categories such as "age" and "sex" are defined based on survey materials / reports of survey institutions / research institutions conducting public opinion surveys, and "corresponding to the categories" are defined. Viewer data relating to “program genre” is accumulated, and the user model processing unit 2
It is managed at 0. In this way, stereotypes (pattern cutting type) are realized by making states and actions recognized for each category. Here, the stereotype refers to the knowledge about stereotypes that we have about groups and their members divided by category, and has been discussed in mutual psychology associated with interpersonal cognition.

The user model managed by the user model processing unit 20 is divided into respective types for each category and managed as a stereotype 21.

In FIG. 2, when the user inquires the program search system about a program, the interface processing unit 11 recognizes it and sends an instruction to the inference engine 13.
The inference engine 13 first acquires the user's basic personal information from the user profile processing unit 14 and the personalized information from the program viewing history, and sends the information to the inference engine 13.

The inference engine 13 sends the user's personal basic information to the user model processing section 20 and specifies the user model by comparing it with the stereotype frame for each category to infer the preference tendency of the program. Since the stereotypes 21 are clearly classified, the user profile processing unit 14 can easily and quickly make predictions and inferences from the input data. If the user profile information is small, for example, if the user is a new user, the search for a new program can be interpolated by referring to a stereotype having similar characteristics.

The program information obtained from the extracted type that is inferred to be related to the user is sent to the inference engine 13, the program is selected by the inference engine 13, and then the program information is output to the interface processing unit 11 to generate an interface. The processing unit 11 edits the display content so that the user can easily recognize it, and displays the edited content on the interface 10. In addition, the system management operation unit 12 controls a series of flows.

By configuring the system as shown in FIG. 2, it is possible to relate to the user from various program information such as the tendency of popularity and novelty of the program obtained from the survey results without specializing in the user profile. It is possible to select and provide certain program information.

Conventionally, a program search system using only the above-mentioned stereotype may be considered. Figure 3
FIG. 1 is a diagram showing an example of the configuration of a conventional program search system using a user model.

FIG. 3 shows an interface 10, an interface processing unit 11, a system management operation unit 12,
The inference engine 13 and the user model processing unit 20 are provided.

When the user inputs a search condition from the interface 10 and searches for a program, the user model processing section extracts the type related to the user and inference engine 1
3 acquires the corresponding program information from the program information database 15 and displays it to the user.

For example, when the search condition is "woman in her twenties", the program information existing in the types "20s" and "female" is acquired, and the inference engine 13 selects the program information as a parameter. , After extracting the program information from the program information database 15, the interface processing unit 11
Displays the program information to the user from the interface 10.

The system shown in FIG. 3 can provide the user with information selected from the statistical data of people who have similar environments such as age and hobbies, and the user can discover potential information. it can.

[0028]

However, all of the above-mentioned program search systems are actively used by the user,
Since information is consciously searched for, if the number of target program information and program contents becomes huge, it will be difficult for the user to find the necessary data and the work will be complicated. .

Therefore, there is a need for a system that actively performs inference activities even for program information such as program genres that the user is not aware of, and serves to discover the subconsciousness of the user. For that purpose, instead of the user performing the work himself / herself, it is necessary for the system to act on behalf of the user to monitor the program information according to his / her preference and the program itself and select and provide only the suitable one.

There is an "agent" as one of the systems acting on behalf of the user. The agent here refers to an artificial system that behaves intelligently (by interaction) like a human being, and uses network distributed processing technology,
It manages elemental technologies such as artificial intelligence, natural language processing technology, voice recognition, and pattern recognition, and autonomously cooperates with each other to build an intelligent system that works as a whole.

Here, when constructing a system using the above-mentioned agent, for example, as shown in FIG. 2, it is divided into a user profile and a user model, and the user model is made to exist as interpolation data of the user profile. When performing processing management with such two systems of databases, the issue is how to allocate and design agents.

Further, in the user model processing section 20,
When changing or adding types, eg managing all types as shown in Figure 3, each type corresponds faithfully to the data and other related types also need to be changed. As a result, the entire system needs to be refurbished, and flexible change work cannot be performed. Therefore, a method of using an agent to update the type can be considered, but in that case, how to arrange an agent that updates data for each stereotype becomes a problem.

The present invention has been made in view of the above-described problems, and a content providing system, a content providing server, and a content providing system in which an agent collects information and can provide program (content) information suitable for a user. It is an object to provide a program and a recording medium recording the program.

[0034]

In order to solve the above problems, the present invention employs means for solving the problems having the following features.

The invention described in claim 1 includes a user,
In a content providing system in which a content provider that provides content to the user and a content providing server that provides content data related to the content of the content provider to the user are connected by a communication network, the content data suitable for the user And an agent for selecting content data suitable for the user using the database, and an agent for providing the content data to the user.

According to the first aspect of the invention, the agent can monitor the program information according to the preference of the user, collect the information, and select and provide only the content suitable for the user.

The invention described in claim 2 is a user,
In a content providing system in which a content provider that provides content to the user and a content providing server that provides content data related to the content of the content provider to the user are connected by a communication network, the content data suitable for the user A database that stores information for selecting the content, an agent that selects content data suitable for the user using the database, and an agent that acquires content data suitable for the selected user from the content provider. It is characterized by having.

According to the second aspect of the present invention, the agent can select the content suitable for the user. In addition, it is not necessary to store the content information provided to the user and the content itself in the system,
It is possible to directly acquire content from a content provider such as a broadcasting station and provide it to the user.

According to a third aspect of the present invention, in a content providing server for providing content data of a content provider to a user via a network, a database storing information for selecting the content data suitable for the user. And one or more agents that select content data suitable for the user using the database and provide the selected content data to the user.

According to the third aspect of the invention, the agent can monitor the program information according to the preference of the user, collect the information, and select and provide only the content suitable for the user.

According to a fourth aspect of the present invention, in a content providing server for providing content data of a content provider to a user via a network, a database storing information for selecting the content data suitable for the user. And one or more agents for selecting content data suitable for the user using the database, acquiring content data suitable for the selected user from the content provider, and providing the user with the content data. Characterize.

According to the fourth aspect of the invention, the agent can select and provide contents suitable for the user. Further, it is not necessary to store the content information to be provided to the user and the content itself in the system, and the content can be directly obtained from the content provider such as a broadcasting station and provided to the user.

According to a fifth aspect of the present invention, the database includes a user profile section for analyzing the user from personal data including personal basic information of the user and content viewing history, and content statistics from a plurality of users. A user model unit for generating and managing a user model for each category from information is provided, and each is managed by an agent.

According to the fifth aspect of the invention, not only the contents specialized for the user but also the contents specialized for the user are obtained by interpolating the information of the model type obtained from the user model unit into the information specialized for the user obtained from the user profile part. It is possible to select and provide content that suits the potential tastes and needs of the.

According to a sixth aspect of the present invention, the user model section has a category management agent that manages a user model for each category, and further has a general agent that collectively manages the agents. Is characterized by.

According to the sixth aspect of the present invention, since the results obtained by a plurality of category agents can be integrated, a user model can be extracted in a timely and accurate manner. Also,
Even if you add types, you can easily deal with them by agents.

The invention described in claim 7 is a content providing server for acquiring content data from a content provider based on a database storing information for selecting content data and providing the content data to the user via a network. In a program for causing a computer to execute provision of content data, a selection agent for selecting content data suitable for the user using the database, and a providing agent for providing the selected content data to the user Causes the computer to execute the processing by.

According to the seventh aspect of the invention, the agent can monitor the program information according to the taste of the user, collect the information, and select and provide only the content suitable for the user.

The invention described in claim 8 is a content providing server for acquiring content data from a content provider based on a database storing information for selecting content data and providing the content data to the user via a network. In a program for causing a computer to execute provision of content data, a process by a selection agent that selects content data suitable for the user using the database, and content data suitable for the selected user The computer is caused to execute the processing by the content acquisition agent acquired from the content provider.

According to the invention described in claim 8, the agent can select the content suitable for the user. In addition, it is not necessary to store the content information provided to the user and the content itself in the system,
It is possible to directly acquire content from a content provider such as a broadcasting station and provide it to the user.

According to a ninth aspect of the present invention, the processing by the selection agent includes a processing by a user profile agent that analyzes the user from personal information including personal basic information of the user and viewing history, and a plurality of users. The contents are selected based on the information obtained from the processing by the user model agent that generates and manages the user model divided for each category from the program statistical information from.

According to the invention described in claim 9, not only the contents specialized for the user but also the contents specialized for the user are obtained by interpolating the information of the model type obtained from the user model part into the information specialized for the user obtained from the user profile part. It is possible to select and provide content that suits the potential tastes and needs of the.

The invention described in claim 10 is characterized in that the processing by the user model agent controls a category management agent that manages a user model for each category.

According to the invention described in claim 10, since the results obtained by a plurality of category agents can be integrated,
An accurate user model can be extracted in a timely manner. Also, even if a type is added, it can be easily dealt with by the agent.

The invention described in claim 11 is the invention according to claim 7.
10 to 10 are stored in a recording medium. A recording medium for recording this program is a magnetic recording medium for magnetically recording information such as a CD-ROM, a flexible disk, or a magneto-optical disk (MO), or an information recording medium such as a ROM or a flash memory. It is possible to use various types of recording media such as a semiconductor memory for recording in a static manner.

[0056]

BEST MODE FOR CARRYING OUT THE INVENTION First, an environment in which a plurality of agents can be activated is prepared in the system by using an agent platform (middleware) which is a base on which agents can operate. Communication between agents is performed using an agent communication language (ACL). The communication act (hereinafter, referred to as CA) included therein is an important element that represents the state or act of communication content, and CA can simulate a conversation between humans such as request, acceptance, refusal, and provision. it can.

For example, when judging whether to accept or reject the data received from the other party,
Since the reliability of the agents and the reliability of the sent message and data are required, a quantified value (using a weighting coefficient) of these is used as the judgment material.

Further, how much each individual agent can grasp (recognize) the user can be evaluated by the calculation result of the data previously stored in the agent and the newly input data. Here, the calculation result may be, for example, one obtained by multiplying the number of corresponding program categories by the points (or probability or occupancy rate) indicated by each category. That is, the conforming amount (fitness)
The greater the number, the more the agent can be perceived by the user as being active and active (both positive and negative aspects).

Next, an embodiment of the invention will be described with reference to the drawings.

FIG. 4 is a diagram showing a first configuration example of the program search system (multi-agent system) according to the present invention.

The program search system shown in FIG. 4 has a human interface agent (hereinafter referred to as HIA) 4
0 and system management agent (hereinafter referred to as MA)
41, an inference engine 42, a user profile agent (hereinafter, UPA) 43, a search agent (hereinafter, SA) 44, a user model agent (hereinafter, UMA) 45, and a stereotype 2
It is composed of 1.

In FIG. 4, by allocating an agent to each processing unit, it is possible to perform ACL communication and to simulate a human conversation such as request, acceptance, refusal, and provision by CA. , It is possible to exchange detailed instructions between each agent.

Further, in FIG. 4, one UMA 45 manages all the plurality of types. Here, the UMA 45 uses an expert system based on production rules to create a database of stereotypes.

With the system configuration shown in FIG. 4, it is not necessary for the user to actively execute the program, and the agent acts on behalf of the user to monitor the program information or the program itself according to his / her preference and select only the appropriate one. Can be provided.

Next, a second configuration example utilizing the present invention will be described with reference to the drawings.

FIG. 5 is a diagram showing a second configuration example of the program search system according to the present invention.

The program search system shown in FIG.
40, MA 41, SA 44, and inference engine 42
, A program information database 15, a UMA 45, and a stereotype agent (hereinafter referred to as STA) 46.

In FIG. 5, S which is in charge of each stereotype
An agent ST that manages stereotypes by installing TA46-1 to STA46-n and giving specific weight to each category.
A46 and an agent UMA45 that controls them and promotes cooperative activities between the agents are installed. STA
By installing 46, each STA46-1 to 46-n
Since the results obtained from can be integrated with UMA45, it is possible to obtain a timely and accurate user model,
In addition, each STA 46 can change each type. Note that the specific gravity here is, for example, weighted in each category so that one user can attach importance to “age” and another user can attach importance to “interest”. , Type 1 ', type 1', and type 1 "are in the same category, but have different specific gravities (probability, occupancy).

Further, the work can be distributed by each STA 46-1 to 46-n sharing the inference that the UMA 45 has collectively performed so far.
The UMA 45 can perform efficient work because it is only necessary to combine the inference results sent from the A46 to derive a comprehensive inference result.

Next, the third structural example including the above-mentioned first structural example and the second structural example, further including an agent for acquiring program information from each broadcasting station, and the flow of processing will be described with reference to the drawings. explain.

FIG. 6 is a diagram showing a third configuration example of the program search system according to the present invention.

In FIG. 6, the program search system includes an interface 10, HIA 40, MA 41, inference engine 42, UPA 43, SA 44, and UMA 4.
5, a STA 46, a program provider agent (hereinafter referred to as PA) 60, and a broadcasting station 61.

Here, the flow of a series of processing in FIG. 6 will be described using the processing flowchart example of FIG.

First, when the user inputs a user name into the interface 10 (S1), the input information is HIA.
Sent to 40. When the information arrives at the HIA 40, it is determined whether the input user is a new user (S2). Here, a database having user information is stored in advance in the HIA 40, and determination is made by referring to it. The determination method is not limited to this.
For example, the UPA 43 may be queried for user information.

In S2, if the user is a new user, the interface 10 is caused to display a screen prompting registration of personal information. The user inputs personal information (S4). If the user is an existing user in S2, it is determined whether or not the personal information already input is changed (S3). If a change instruction is given in S3, the personal information is changed (input) from the personal information input screen (S
4) is performed.

If no change is made in S3, or S4
When personal information is input in, the HIA 40 determines that each agent can understand the communication language between agents (AC
L) is converted (S5). The data converted in S5 is sent to SA44.

The SA 44 obtains the program information related to the user from the user profile and the user model,
The personal data of the user is sent to the UPA 43 and UMA 45 (S6).

The UPA 43 extracts user-specific data from the input data based on the viewing history of the user in the user profile and sends the extracted related information to the SA 44 (S7). In addition, UMA45 is for each ST
The personal data is sent to A46, and the sent STA46 collates its own stereotype with the input data, and sends the fitness calculated from the specific gravity of each STA to the UMA45. The UMA 45 calculates the fitness obtained from each STA 46 and sends the type of information most suitable for the user to the SA 44 (S8).

The SA 44 derives the program recommendation information (for example, program genre, etc.) inferred to be suitable for the user sent from the UPA 43 and UMA 45 as a selection (search) result (S9). Next, the selection result obtained in S9 is sent to the PA 60, the PA 60 inquires of each broadcasting station 61 about the program information suitable for the selection result from SA 44, and after receiving the required program from the broadcasting station 61, the program The program description information (EPG) describing the information and the content of the program is sent to SA44 (S10). At SA44, the information sent from PA60 is converted into a format that can be provided to the user, and the information is converted into H format.
Send to IA40 (S11). SA44 in HIA40
The program information sent from the device is converted into a format that is easy for the user to view and provided to the user via the interface 10 (S
12) Do.

Further, when the HIA 40 recognizes the program which the user has actually instructed to view from the provided programs and sends the program information to the UPA 43 and the UMA 45, the agents of the two sides reflect the program information in the database (S1).
3) Do.

Finally, it is determined whether or not there is a user change from the interface (S14). If not, the process is terminated, and if there is, the process returns to step S1.

By the procedure described above, not only related program information specialized for the user can be obtained, but also a program inferred that is potentially suitable for the taste and needs can be newly provided. From the type, you can grasp trend programs of the same generation and information that people with similar hobbies are interested in, so you can have a deeper interest in the provided program information and meet the needs of the program. You can deepen. Further, by providing the PA 60, it is not necessary to store various program information in the system, and the latest information can be obtained from each broadcasting station.

Next, the configuration of each device terminal of the program search system shown in FIG. 6 and the specific processing flow will be described with reference to the drawings.

FIG. 8 is a diagram showing an example of the configuration of each terminal device of the program search system utilizing the present invention.

In FIG. 8, the interface 10, the mediation system (MA) 41, the program search (SA) 44, and the program provider (PA) 60 are constructed on the same network by the network interface.

The interface 10 has the function of the HIA 40 and is a programmed main controller (control means, hereinafter referred to as CPU) 53 for controlling the entire input / output of the system.
An input device 5 including a pointing device such as a keyboard and a mouse via a human interface
0, a display device 51 used as a monitor for displaying an execution screen, an input result, etc., and an output device 5 for outputting a searched program.
2 is connected.

SA44 is UPA43, UMA4
5, a CPU 54 programmed with the function of the inference engine 42, which controls the entire SA 44, a storage device 56 in which a user profile is stored, a storage device 58 in which a user model is stored, and a processing state monitor in the SA 44. A display device 57 to be used and an output device 55 for outputting the retrieved and inferred program information are connected.

Further, the PA 60 extracts program information (including the program itself) corresponding to the inquiry about the program information from the SA 44 from the broadcasting station 61, and adds program description information (EPG) to the SA 44. To the programmed CPU 59 that controls the entire sending function to the PA6.
A display device 71 that displays a processing state of 0, an output device 72 that outputs a processing result, a program information receiving device 73 that receives program information from the broadcasting station 61, and a television receiving device 74 that receives a television signal from the broadcasting station 61. A storage device 75 for storing program description information is connected.

Further, the MA 41 has a display for displaying the processing state of the MA 41 on the programmed CPU 76 which controls the entire multi-agent program search system such as the processing of each agent connected to the network and the data flow. The device 77 and the output device 78 for outputting the processing result are connected. The device configuration of each agent is not limited to this.

Next, the display device 5 of the interface 10
The flow of the process will be specifically described based on the screen transition example displayed in 1.

FIG. 9 is a diagram showing a transition example of the screen displayed on the interface when the system is implemented.

First, the CPU 53 causes the display device 51 to display the image shown in FIG.
A login screen 90 as shown in (a) is displayed. The login screen 90 is configured to include a user name input area 101, a “stereotype registration / change” button 102, a “search start” button 103, and a “cancel” button 104.

The user inputs the user name from the input device 50 into the user name input area 101. The method of inputting the user name is not limited to this. For example, the interface 10 may be provided with an agent in charge of natural language processing so that the user name can be recognized in a natural language, and the character can be operated. It is also possible to provide a helper agent that enables the input of the user name to be confirmed using the character. (Example: “Hanako Tanaka?”) Next, the CPU 53 determines whether or not the input user name is a new user, and in the case of a new user, the display device 51.
Display a message prompting you to register. (Example: Please register as a new user.) When the “register / change stereotype” button 102 on the login screen 90 is pressed, the CPU 53 displays the stereotype registration screen on the display device 51. An example of a stereotype registration change screen is shown in FIG.

The stereotype registration change screen 91 is displayed on the user name input area 101 and the current stereotype area 10.
5, personal information input area 106, "register" button 107,
"Clear" button 108 and "Cancel" button 1
04 and 04.

The user inputs personal information (stereotype information) such as family structure and interests on the stereotype registration screen 91 provided by the CPU 53. (Example: Address: 999-99, Setagaya-ku, Tokyo, age group: 20s, interest: education) After input, press the "Register" button 107. It should be noted that by pressing the “clear” button 108 while inputting the user registration, the data on the screen that has been input so far can be deleted, and by pressing the “cancel” button 104, the registration process can be completed. The process can be terminated.

When the information is input, the CPU 53 causes the HI
Since it has the function of A40, the HIA 40 extracts necessary information from the registration information and converts the information into the inter-agent communication language (ACL). (Example: (firstname Hanako) (lastname Tanaka) (age: 20) (interests education)) When the conversion is completed, the CPU 53 transmits data to the SA 44 via the network.

Upon receiving the data transmitted from the interface 10, the CPU 54 of SA44 receives the data from the UPA 43.
Since it has the functions of the UMA 45 and the UMA 45, the registration information is reflected in the user profile information of the storage device 56 and the stereotype information of the user model of the storage device 58 by sending data to each.

Here, examples of databases in the storage device 56 and the storage device 58 will be described with reference to the drawings.

FIG. 10 is a diagram showing an example of a user profile.

FIG. 10A shows the program history information of the user, which includes items such as user name, program ID, viewing date, and day of the week viewed, and FIG. 10B shows the user name, occupation, and
It is the user's personal basic information composed of items such as interest and age.

The user profile collectively manages history information of programs watched by each user, and program information specialized for the user from accumulated information and basic personal information (for example, user's favorite program genre). Etc. can be extracted.

Next, the user model will be described. First,
First, the user inputs information as shown in FIG. 10B as stereotype registration or change information.

The UMA 45 reflects the data in each stereotype based on the user information shown in FIG. 10 (b). Here, FIG. 11 shows an example of the occupation rate (specific gravity) of each stereotype in each category. FIG. 11A shows the occupancy rate of the category “gender”, FIG. 11B shows the occupancy rate of the category “age”, and FIG. 11C shows the occupancy rate of the category “interest”. Indicates the occupancy rate.

A table as shown in FIG. 11 has STAs 46-1 to 4-4.
It exists corresponding to 6-n, and when extracting the user model,
The STAs 46-1 to 46-n collate the input data with the held model to calculate the fitness.

Here, FIG. 12 shows an example of each stereotype index for each user. By providing a table as shown in FIG. 12 in the database, the user can easily understand which stereotype and high adaptability.

Next, by inputting the user name on the login screen 90 and pressing the "search start" button 103, the CP
U53 generates an ACL and sends a search instruction to SA44.

When the SA 44 receives the data transmitted from the interface 10, the CPU 54 makes the UPA 4
3 and the UMA 45 function, data is sent to each of them, user-specific information is extracted from the user profile information of the storage device 56, and viewer statistics are obtained from the stereotype information of the user model of the storage device 58. The user model related to the input data is extracted based on the target information, and each extracted data is sent to the CPU 54. (Example: UP
The A43 sends the program history information that Mr. Tanaka has so far, and the UMA 45 sends to the CPU 54 the user model information that a woman in her twenties often sees. The information here is the program genre (movie,
Sports, etc.) and their fitness, stereotype index, etc. are mainly used. ) The CPU 54 uses the information obtained above to set the PA 60.
Ask to provide program information to. PA60 is CPU59
Upon reception, the broadcasting station 61 having the corresponding program information is inquired, and the program information receiving device 73 and the television receiving device 74 acquire the program information and the program itself and store them in the storage device 75. Further, the CPU 59 generates program description information (EPG) from the acquired program information and stores it in the storage device 7.
Store in 5. Here, the program description information (EPG) will be described with reference to the drawings.

FIG. 13 is a diagram showing an example of the program description information.

The program description information shown in FIG. 13 is made up of items such as a program ID, a program genre, a program name, and a program net time. The CPU 59 transmits the program description information obtained from each broadcasting station 61 to the SA 44 via the network.

The program information acquiring means is not limited to this, and a frequently used program may be stored in the storage device 75, or even if one broadcasting station itself performs the processing of PA60. Good. Further, the processing state in the PA 60 can be always confirmed on the display device 71, and if the processing state is desired to be output, it can be output by the output device 72.

Upon receiving the program description information from the PA 60, the CPU 54 performs processing such as rearranging the program description information in descending order of interest, and then transmits the program description information to the interface 10 via the network.

The series of processing states of SA44 can always be confirmed on the display device 57, and the processing states can be output using the output device 55.

When the CPU 53 receives the search result from SA44, it generates a program list as shown in FIG. 9C,
It is displayed to the user from the display device 51. The user selects a program he / she wants to watch from the displayed program list screen 92 and presses the “start” button 110 to display the CP
The U53 requests the program instructed by the PA 60 and at the same time informs the user that the program instructed by the SA 44 has been viewed. The CPU 59 transmits the corresponding program to the interface 10, and the CPU 54 reflects the viewed program and user information on the user profile of the storage device 56 and the stereotype of the storage device 58.

The CPU 53 causes the output device 52 to display the program received from the PA 60.

Finally, the end screen 93 shown in FIG. 9D is displayed. If the "end" button is pressed, a series of processing is ended, and if the "change user" button 112 is pressed, login is performed. The process returns to the screen 90 and the process is continued.

Further, the CPU 76 in the MA 41 controls the communication of data between the interface 10, the SA 44, and the PA 60, at the same time manages each agent, and the monitoring state can be constantly confirmed on the display device 77. The output device 78 can output the monitoring status.

By the processing procedure described above, a program suitable for the user can be searched and provided. Various users can use the system of the present invention to construct detailed stereotypes, and more profile data can be accumulated as the number of times the user uses it increases. We can provide potential potential programs and programs that correspond to new environments and trends of the same generation.

According to the embodiment of the present invention, the user
By using the content providing system, the content providing server, the program, and the recording medium recording the program, the program (content) specialized for the user is interpolated, and the user is not restricted by his / her registered profile. You can pick up new programs that are potentially suitable for your tastes and needs.

Further, by dividing the user model into agents for each stereotype and changing the specific gravity of each category of each stereotype, it is possible to obtain a timely and accurate user model from various viewpoints. Also, since each agent manages the update of each stereotype, it is easy to update the statistical data corresponding to the new environment and trends obtained from opinion polls, etc. Can be easily confirmed and updated.

Furthermore, by incorporating a learning function and a feedback function in the user profile / user model selection function, it is possible to support program inference and program selection from the next time onward. The embodiment of the present invention is not limited to this, and various modifications can be carried out within the technical scope thereof.

[0121]

According to the present invention, it is possible to provide a content providing system for providing content suitable for a user, a content providing server, a program therefor, and a recording medium recording the program.

[Brief description of drawings]

FIG. 1 is a diagram of a first configuration example of a conventional program search system.

FIG. 2 is a diagram of a second configuration example of a conventional program search system.

FIG. 3 is a diagram showing a configuration example of a conventional program search system using a user model.

FIG. 4 is a diagram of a first configuration example of a program search system (multi-agent system) according to the present invention.

FIG. 5 is a diagram showing a second configuration example of the program search system in the present invention.

FIG. 6 is a diagram showing a third configuration example of the program search system according to the present invention.

FIG. 7 is an example of a processing flowchart based on a third configuration example.

FIG. 8 is a diagram showing a configuration example of each terminal device of the program search system using the present invention.

FIG. 9 is a diagram showing a transition example of a screen displayed on the interface when the system is implemented.

FIG. 10 is a diagram showing an example of a user profile.

FIG. 11 is a diagram showing an example of occupation ratio (specific gravity) of each stereotype in each category.

FIG. 12 is a diagram of an example of each stereotype index for each user.

FIG. 13 is a diagram showing an example of program description information.

[Explanation of symbols]

10 Interface 13, 42 Inference Engine 40 Human Interface Agent (HI
A) 41 system management agent (MA) 43 user profile agent (UPA) 44 search agent (SA) 45 user model agent (UMA) 46 stereotype agent (STA) 60 program provider agent (PA)

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G06F 17/60 ZEC G06F 17/60 ZEC (72) Inventor Masahiro Shibata 1-10-11 Kinuta, Setagaya-ku, Tokyo Issue Japan Broadcasting Corporation F-Term in Broadcasting Technology Laboratory (reference) 5B075 ND20 NS01 PR03

Claims (11)

[Claims] 1. A content providing system in which a user, a content provider that provides content to the user, and a content providing server that provides content data related to the content of the content provider to the user are connected by a communication network. A database for storing information for selecting the content data suitable for the user, an agent for selecting the content data suitable for the user using the database, and an agent for providing the content data to the user. A content providing system characterized by having. 2. A content providing system in which a user, a content provider that provides content to the user, and a content providing server that provides content data relating to the content of the content provider to the user are connected by a communication network. A database for storing information for selecting the content data suitable for the user, an agent for selecting content data suitable for the user using the database, content data suitable for the selected user A content providing system comprising: an agent obtained from a provider. 3. A content providing server for providing content data of a content provider to a user via a network, a database storing information for selecting the content data suitable for the user, and the database using the database. A content providing server, comprising: one or a plurality of agents that select content data suitable for a user and provide the selected content data to the user. 4. A content providing server for providing content data of a content provider to a user via a network, a database storing information for selecting the content data suitable for the user, and the database using the database. A content providing server comprising: one or a plurality of agents for selecting content data suitable for a user, acquiring content data suitable for the selected user from the content provider, and providing the user with the agent. 5. The database includes a user profile section for analyzing a user from personal data including personal basic information of the user and a content viewing history, and a user model for each category based on content statistical information from a plurality of users. The content providing server according to claim 3 or 4, further comprising a user model unit that generates and manages each of which is managed by an agent. 6. The user model section has a category management agent that manages a user model for each category, and further has a general agent that collectively manages the agent. Content providing server. 7. An agent is provided for providing content data in a content providing server which obtains content data from a content provider based on a database storing information for selecting content data and provides the user with the content data via a network. In a program to be executed by a computer, the computer is caused to execute a selection agent for selecting content data suitable for the user using the database, and a process by a providing agent for providing the selected content data to the user. Program for. 8. An agent is provided for providing content data in a content providing server which obtains content data from a content provider based on a database storing information for selecting content data and provides the user with the content data via a network. In a program to be executed by a computer, processing by a selection agent for selecting content data suitable for the user using the database, and content for acquiring content data suitable for the selected user from the content provider A program that causes a computer to perform processing by an acquisition agent. 9. The processing by the selection agent includes processing by a user profile agent that analyzes the user from personal information including personal basic information and viewing history of the user, and program category information from a plurality of users for each category. 9. The program according to claim 7, wherein the content is selected based on information obtained from the processing by the user model agent that generates and manages the user model divided into. 10. The processing by the user model agent controls a category management agent that manages a user model for each category.
The listed program. 11. A recording medium storing the program according to claim 7.