JP2006285299A - Ability filter type content recommendation device, method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an environment enabling a user's efficient browsing by recommending a providing content according to difficulty level of content and user's ability. <P>SOLUTION: A difficulty level estimation processing part 23 and an ability estimation processing part 24 perform estimation of the difficulty level of content and estimation of the user's ability, respectively, and an ability filter processing part 21 performs filtering of contents based on the user's ability. An evaluation data fetching part 22 collects users' evaluation for contents provided, and updates a content information DB 230 based on the user's ability and an evaluation value for content of the user concerned, and updates a user ability information DB 240 based on this result. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a capability filter type content recommendation device, method, and program for recommending provided content in accordance with the degree of difficulty of content and user capability.

Many techniques for improving the efficiency of information retrieval and providing optimal content to users are known. Specifically, techniques for providing recommended content based on user preferences and search history have been proposed (see, for example, Patent Documents 1, 2, and 3).
JP 2004-78689 A JP 2003-224798 A JP 2005-33600 A

  However, according to the above-described prior art, for example, there is a difference in reading comprehension of kanji and English between the lower grades of elementary school and university students. There is a problem that it is not always possible content. In addition, since there is a dependency relationship between the difficulty level of the content and the user's ability, there is a problem that proper content cannot be obtained simply by replacing the preference filter system data with the difficulty level of the content and the user's ability. there were.

  The present invention has been made based on the above circumstances, and recommends an optimum provided content according to a user's ability and can provide an efficient browsing environment, an ability filter type content recommendation apparatus, method, and program The purpose is to provide.

  In order to solve the above-described problems, the present invention has a capability having a content information database storing content and information indicating the difficulty level of the content, and a user capability information database storing information indicating the capability of the user. A filter type content recommendation device that receives a browsing request from the outside, sequentially acquires content from the content information database, information indicating the difficulty level of the acquired content, and a user stored in the user capability information database The information indicating the ability of the user is compared, and when the browsing is permitted, the ability filtering processing unit that provides the acquired content to the outside, and the user's evaluation of the provided content are collected, and the collected evaluation The degree of difficulty of the content is estimated from the information indicating the ability of the user and the browsing history. Then, a new user ability value is estimated from the difficulty level estimation processing unit stored in the content information database and information indicating the estimated difficulty level of the content and the user ability, and stored in the content information database. And an ability estimation processing unit.

  In the present invention, the capability filter processing unit compares the difficulty level of the content stored in the content information database with the capability value of the user stored in the capability information database of the user, and the capability value of the user It is characterized by providing content with a difficulty level not exceeding.

  Further, in the present invention, at least a part of the content provided to the user is given an evaluation option for the user to explicitly give an evaluation of the content, and the selection input by the user who has browsed the content is taken in, An evaluation data capturing unit that is reflected in estimation of the difficulty level by the difficulty level estimation processing unit.

  In the present invention, a dictionary that can refer to the meaning or reading of each Chinese character or word having a high degree of difficulty in the content is assigned to the content, and the number of times the user has referred to the dictionary is captured. And an evaluation data capturing unit to be reflected in the estimation of the difficulty level by the degree determination processing unit.

  Further, in the present invention, the difficulty level estimation processing unit determines an initial difficulty level preset for each content in the content information database according to a predetermined determination criterion based on data captured by the evaluation data capturing unit. It is characterized by updating.

  In the present invention, the capability estimation processing unit updates the capability value set in advance in the capability information database for each user according to a predetermined criterion based on the data captured by the evaluation data capturing unit. It is characterized by.

  The present invention further includes a content information database storing content and information indicating the difficulty level of the content, and a user capability information database storing information indicating the capability of the user. A content filtering method that recommends the content provided according to the information received from the outside, receives the browsing request from the outside, sequentially acquires the content from the content information database, information indicating the difficulty of the acquired content, Comparing information indicating the user's ability stored in the user ability information database, providing the acquired content to the outside when browsing is permitted, and collecting user evaluations on the provided content The collected evaluation, information indicating the user's ability, and browsing history The content difficulty level of the content is estimated and stored in the content information database, and a new user ability value is estimated from the estimated content difficulty level and information indicating the user ability. Storing in an information database.

  The present invention also relates to a capability filter type content recommendation device having a content information database storing content and information indicating the difficulty level of the content, and a user capability information database storing information indicating the capability of the user. Information used to receive a browsing request from outside, sequentially acquire content from the content information database, information indicating the difficulty level of the acquired content, and user capabilities stored in the user capability information database And collecting the user's evaluation of the provided content when the browsing is permitted, collecting the user's evaluation of the provided content, and the user's ability The degree of difficulty of the content is estimated from the information indicating the browsing history and the browsing history, and the content A process for storing in the content information database, and a process for estimating a new user ability value from the estimated content difficulty level and information indicating the user ability and storing in the content information database. It is made to perform.

According to the present invention, the difficulty level estimation processing unit and the capability estimation processing unit estimate the difficulty level of the content and the user capability, respectively, and the capability filter processing unit filters the content based on the user capability (for the user's capability). (Recommendation of the provided content), it is possible to recommend the optimum provided content according to the reading level of the user. In addition, the evaluation data capturing unit collects user evaluations for the provided content, thereby updating the content information database based on the user capability and the evaluation value for the user content. By updating, it is possible to maintain the difficulty level of the content and the ability value of the user at appropriate values.
As a result, the user can browse only readable content, and efficient browsing is possible. Moreover, it can also set so that it may recommend also the thing whose difficulty level is a little over a user's ability value, In this case, it is applicable also to the learning system aiming at raising a user's ability value.

FIG. 1 is a block diagram showing a basic configuration of a capability filter type content recommendation device according to an embodiment of the present invention.
The capability filter type content recommendation device of the present invention may be implemented as a stand-alone configuration or in a server connected to a network with a user terminal. Here, the latter is shown as an embodiment.

The capability filter type content recommendation device 2 according to the embodiment of the present invention is implemented in a server, and includes a capability filter processing unit 21, an evaluation data capturing unit 22, a difficulty level estimation processing unit 23, and a capability estimation processing unit 24. It consists of. The user terminal 1 is connected via a network.
The capability filter processing unit 21 receives a request (browsing request) from the user terminal 1, sequentially acquires content from a content information DB (Data Base) 230 included in the difficulty level estimation processing unit 23, and the difficulty level of the acquired content Is compared with information indicating the user's ability stored in the user ability information DB 240 of the ability estimation processing unit 24, and when browsing is permitted, the previously acquired content is provided to the user terminal 1. It has a function to do.

On the other hand, the difficulty level estimation processing unit 23 collects the user's evaluation of the provided content via the evaluation data capturing unit 22, and determines the difficulty of the content from the collected evaluation, information indicating the user's ability, and the browsing history. It has a function of estimating the degree and storing it in the content information DB 230.
The capability estimation processing unit 24 has a function of estimating a new user capability value from the estimated content difficulty level and information indicating the user capability, and storing it in the content information DB 230.
In addition, the evaluation data capturing unit 22 assigns an evaluation option for the user to explicitly give an evaluation to the content to at least a part of the content provided to the user, and captures a selection input by the user who has viewed the content Alternatively, a dictionary that can refer to the meaning or reading of each kanji or word having a high degree of difficulty in the content is assigned to the content, the number of times the user has referred to the dictionary is taken in, and the difficulty level determination processing unit 23 It has a function to reflect in the estimation of the difficulty level. In addition, regarding the input of evaluation for content, a method using a character input field that can be freely input may be used.

FIG. 2 is a flowchart quoted for explaining the basic operation of the capability filter type content recommendation device according to the embodiment of the present invention.
The basic operation of the capability filter type content recommendation apparatus shown in FIG. 1 will be described below with reference to the flowchart shown in FIG.

First, a user operates the user terminal 1 to connect to a server (capability filter type content recommendation device 2) via a network, and issues a browsing request.
In the capability filter type content recommendation device 2, the capability filter processing unit 21 accepts this browsing request (S21). The capability filter processing unit 21 acquires content one by one from the content information DB 230 (S22), compares the difficulty level of the content with the capability value of the user, and as a result, determines whether or not browsing permission can be determined. (S23). Details of the data structure of the content information DB 230 will be described later.
As a result of the browsing determination (S24), if browsing is permitted, the content is targeted for the provided content (S25), and if not, the content is not targeted (S26). Then, the presence / absence of the following content is checked (S27), and the same processing after S22 is repeated in order to read the next content and perform the same processing as above (S28). The basic operation ends when there is no more content.

Next, the operation of the capability filter type content recommendation device according to the embodiment of the present invention will be described with reference to the flowchart showing the flow of content difficulty level and user capability value update processing shown in FIG.
Here, the evaluation data capturing unit 22 collects the evaluation of the provided content from the user via the user terminal 1 and supplies it to the difficulty level estimation processing unit 23 (S31). A mechanism for obtaining an evaluation from the user will be described later. The difficulty level estimation processing unit 23 estimates the difficulty level of the content based on the user evaluation captured via the evaluation data capturing unit 22 and the user capability value stored in the user capability information DB 240 (S32). Then, the estimated difficulty level is stored in the content information DB 230 (S33). The logic for difficulty level estimation by the difficulty level estimation processing unit 23 will be described later.

  On the other hand, the capability estimation processing unit 24 estimates a new capability value from the difficulty level of the given content, the capability of the user, and the browsing history (S34), and stores this capability estimation result in the content information DB 230 (S35). ). The data structure of the user capability information DB 240 and the logic for capability estimation by the capability estimation processing unit 24 will be described later.

FIG. 4 is a block diagram showing a detailed internal configuration of the capability filter type content recommendation device according to the embodiment of the present invention.
In FIG. 4, blocks denoted by the same reference numerals as those shown in FIG. 1 have the same names and functions as the blocks shown in FIG. Reference numeral 25 denotes a request reception unit that receives a browsing request from the user terminal 1 and supplies the request to the capability filter processing unit 21. Reference numeral 26 denotes a providing unit, which serves as an interface for providing content recommended through the capability filter processing unit 21 to the requested user terminal 1. Reference numeral 27 denotes a user registration unit, which serves as an interface for registering capability values and the like in the user capability information DB 240 at the time of initial setting described later. Reference numeral 28 denotes an initial value setting unit which serves as an interface for registering the difficulty level of the content in the content information DB 230.

The detailed operation of the capability filter type content recommendation apparatus according to the embodiment of the present invention shown in FIG. 4 will be described below.
First, the flow of initial setting to the user capability information DB 240 and the flow of initial setting to the content information DB 230 at the time of user registration will be described with reference to the initial setting flowcharts shown in FIGS.

In FIG. 5, when a user inputs various data for user registration, such as an ID, via the user terminal 1, the user registration unit 27 of the capability filter type content recommendation device 2 captures these data (S51: user registration). The initial value is written in the user capability information DB 240 (S52). An example of the data structure of the user capability information DB 240 is shown in FIG.
As shown in FIG. 10A, the user capability information DB 240 includes item data of a user ID, a capability value vector, and a browsing history. Here, the “user ID” is information that can uniquely identify the user, and the “ability value vector” is one or more ability values, such as “communication”, “medical”, etc. A capability value may be assigned for each field. Also, “browsing history” refers to a list of contents viewed by the user.

In FIG. 6, the initial value setting unit 28 of the capability filter type content recommendation device 2 takes in the input content ID for each content (S61: initial setting) and writes it as an initial value in the content information DB 230 (S62). An example of the data structure of the content information DB 230 is shown in FIG.
As shown in FIG. 10B, the content information DB 230 includes item data of a content ID, a difficulty level vector, content details, and an evaluation history. Here, the “content ID” is information that can uniquely identify the content, and the “difficulty vector” means one or more difficulty levels. For example, “communication”, “medical”, etc. A difficulty level may be assigned to. The “content” refers to information such as a title, text, image, and update time, and the evaluation history refers to a user's evaluation history.

In the above configuration, when the capability filter processing unit 21 receives a browsing request from the user terminal 1 via the request receiving unit 25, the difficulty level of the content included in the content information DB 230 and the user capability included in the user capability information DB 240. Compare the value. As a result of the comparison, for example, only content having a difficulty level that does not exceed the user's ability value is provided as a recommended content to the requested user terminal 1 via the providing unit 26.
In the case where the main purpose is to improve the user's ability, the ability filter processing unit 21 permits it even if it slightly exceeds the ability value of the user. Further, even if the capability filter processing unit 21 also has a content filter function using conventional preference information, the essential function of the capability filter type content recommendation device 2 of the present invention is not affected at all and can be linked without any problem. For example, when linked with a content filter using preference information, only the content that does not exceed the user's ability and is of interest to the user can be browsed.

Further, the evaluation data capturing unit 22 can capture an evaluation input from a user who has browsed the provided content, and can make it back to the difficulty level estimation by the difficulty level estimation processing unit 23 and reflect it. An implementation example of feedback and an implementation example of implicit feedback will be described.
As an example of realizing the former explicit feedback, an evaluation option is given to a part of the content provided to the user so that the user gives an evaluation to the explicit content. After browsing the content, the user selects one of these options and transmits data for evaluation. FIG. 7 shows an example of the screen configuration. Here, the user selects one of the three options “easy”, “normal”, and “difficult”. In addition, regarding the input of evaluation for content, a method using a character input field that can be freely input may be used.

As an implementation example of the latter implicit feedback, a dictionary reference mechanism that can easily refer to meanings and readings is given to contents with respect to kanji and words having high difficulty in the contents. Here, feedback is realized by evaluating the number of times the user has referred to the dictionary. FIG. 8 shows an example of the screen configuration.
Here, we adopted a mechanism that pops up a window showing the reading kana when the user moves the mouse or cursor to a difficult word.

On the other hand, the difficulty level estimation processing unit 23 sets the abilities of the two users A and B to “high” and “low”, and updates the difficulty level of the content according to the user's evaluation. If the evaluation given by each user to the provided content is made into two stages of “difficult” and “easy”, for example, the difficulty level of the content is determined according to the determination criteria shown in Table 1 in FIG. decide.
A specific description will be given below. Temporarily, the difficulty levels of the two contents are set to “high” and “low”, and the user ability is updated according to the evaluation given to each user by the provided contents. If the evaluation given to each provided content by the user is made into two stages of “difficult” and “easy”, the user's ability is determined by the determination as shown in <Table 2> in FIG. 9B. .

Next, an initial value setting method related to the difficulty level of content will be described. First, the difficulty level of an article is calculated based on character information included in the content. Also, the difficulty level of content related to Japanese and English is calculated according to the user's ability and grade.
The calculation target is Japanese content for the kanji level. The target grades are “1st grade” to “6th grade” and “more than that”. Here, the weighted average shown by the following arithmetic expression (1) is calculated, and this is set as the difficulty level.
Σ _Grade (School Year * Kanji Appearance Type / Total Kanji Appearance Type) ... (1)
FIG. 11 shows an example of difficulty level initial value setting for Japanese content.

On the other hand, for the English level, the calculation target is English content. The target grades are “junior high school year 1” to “junior high school year 3”, “high school year 1” to “high school year 3”, and “more than that”. Here, a skimming process is performed to correct the stem. Also, stop words are removed. Then, a weighted average is calculated according to the following arithmetic expression (2), and this is set as the difficulty level.
Σ _Grade (School Year * Number of Word Appearance Types / Total Number of Word Appearance Types) (2)
The calculation target is not limited to English content, and may be content in other foreign languages.

FIG. 12 is a flowchart showing the flow of processing for calculating the initial difficulty level for the content described above.
In FIG. 12, based on a request from the user, the initial value setting unit 28 performs a calculation for calculating a difficulty initial value every time content is extracted (S121). Specifically, the type of content is determined (S122). If the extracted content is Japanese, the above equation (1) is used, and if it is English, the above equation (2) is used. Each degree is calculated (S123, S124), and an initial value as a difficulty level to be stored in the content information DB 230 is determined (S125).

FIG. 13 shows an example of an initial setting related to the ability value when user registration is performed in a table format.
FIG. 13 (a) shows the case of kanji, <Table 3> shows an example of the user's initial reading comprehension ability, and FIG. 13 (b) shows the case of English as an example of the user's initial English reading ability. The value set here is stored in the user capability information DB 240.

14 and 15 are block diagrams cited for explaining an application example of the capability filter type content recommendation device of the present invention.
FIG. 14 is a configuration example of a content recommendation system in which a dictionary reference function is combined with the ability filter type content recommendation device of the present invention. According to FIG. 14, the dictionary reference unit 3 can realize a service that assists the user in understanding by assigning a kana or a word meaning to the content in accordance with the difficulty level and the ability value of the user. Here, in addition to the dictionary reference unit 3, additional information DB 4 corresponding to a dictionary indicating the meaning of kanji reading and English words is required.

  FIG. 15 is a functional configuration example of a content recommendation system in which a preference information determination function is combined with the ability filter type content recommendation device of the present invention. According to FIG. 15, by performing filtering based on preference information in addition to filtering based on the user's ability, it becomes possible to browse only the content that the user can understand and is interested in, which is more efficient. Browsing service can be realized. Here, a user preference information DB 6 is added, and a preference information determination unit 5 that refers to this and reflects the preference information in filtering is added.

  As described above, according to the present invention, the difficulty level estimation processing unit 23 and the capability estimation processing unit 24 estimate the difficulty level of the content and the user capability, respectively, and the capability filter processing unit 21 determines the content based on the user capability. By performing the filtering, it is possible to recommend the optimum provided content according to the reading level of the user. Further, the evaluation data capturing unit 22 collects user evaluations for the provided content, thereby updating the content information DB 230 based on the user capability and the evaluation value for the user content, and based on the result, the user capability By updating the information DB 240, the difficulty level of the content and the ability value of the user are maintained at appropriate values.

  As a result, the user can browse only readable content, and efficient browsing is possible. In addition, it is possible to set so that the difficulty level slightly exceeds the ability value of the user, and in this case, the present invention can also be applied to a learning system for increasing the ability value of the user. In addition, since it is not necessary to generate and store content for each user on the server side, resource consumption does not increase, and a capability filter type content recommendation mechanism can be efficiently constructed.

  1 and 4, the capability filter processing unit 21, the difficulty level estimation processing unit 22, the capability estimation processing unit 23, the evaluation data capturing unit 24, the request receiving unit 25, the providing unit 26, the user registration unit 27, and the initial stage Procedures executed by each of the value setting units 28 are recorded on a computer-readable recording medium, and a program recorded on the recording medium is read into a computer system and executed, thereby executing the capability filter type content recommendation device of the present invention. Can be realized. The computer system here includes an OS and hardware such as peripheral devices.

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

  The embodiments of the present invention have been described in detail with reference to the drawings. However, the specific configuration is not limited to the embodiments, and includes designs and the like that do not depart from the gist of the present invention.

It is a block diagram which shows the basic composition of the capability filter type content recommendation apparatus which concerns on embodiment of this invention. It is a flowchart which shows the basic operation | movement of the capability filter type content recommendation apparatus which concerns on embodiment of this invention. 5 is a flowchart cited for explaining the operation of the capability filter type content recommendation device according to the embodiment of the present invention. It is a block diagram which shows the specific structure of the capability filter type content recommendation apparatus which concerns on embodiment of this invention. 6 is a flowchart cited for explaining an initial setting operation of the capability filter type content recommendation device according to the embodiment of the present invention. 6 is a flowchart cited for explaining an initial setting operation of the capability filter type content recommendation device according to the embodiment of the present invention. It is a figure which shows an example of the screen structure used in embodiment of this invention. It is a figure which shows an example of the screen structure used in embodiment of this invention. It is the figure which showed the difficulty level of the content used in embodiment of this invention, and the user's ability determination example in a table format. It is a figure which shows an example of the data structure of the database used by embodiment of this invention. It is a figure which shows an example of the difficulty initial value of a Japanese content used in embodiment of this invention. It is a flowchart which shows the flow of a process about content difficulty level initial value calculation in embodiment of this invention. It is the figure which showed the example of the capability value initial setting of a user used in embodiment of this invention in a table format. It is a block diagram which shows the application example of the capability filter type | mold content recommendation apparatus which concerns on embodiment of this invention. It is a block diagram which shows the application example of the capability filter type | mold content recommendation apparatus which concerns on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... User terminal, 2 ... Capability filter type content recommendation apparatus, 21 ... Capability filter processing part, 22 ... Difficulty degree estimation processing part, 23 ... Capability estimation processing part, 24 ... Evaluation data taking-in part, 25 ... Request reception part, 26 ... Providing section, 27 ... User registration section, 28 ... Initial value setting section, 230 ... Content information DB, 240 ... User capability information DB

Claims (8)

A capability filter type content recommendation device having a content information database in which information indicating content and the degree of difficulty of the content is stored, and a user capability information database in which information indicating the capability of the user is stored,
A browsing request is received from the outside, content is sequentially acquired from the content information database, and information indicating the difficulty level of the acquired content is compared with information indicating the ability of the user stored in the user capability information database. An ability filter processing unit for providing the acquired content to the outside when browsing is permitted;
The user's evaluation for the provided content is collected, the difficulty level of the content is estimated from the collected evaluation, information indicating the user's ability, and browsing history, and stored in the content information database A processing unit;
A capability estimation processing unit that estimates a new user capability value from the estimated content difficulty level and information indicating the user capability, and stores the estimated value in the content information database;
A capability filter type content recommendation device comprising: The capability filter processing unit
The difficulty level of the content stored in the content information database is compared with the capability value of the user stored in the capability information database of the user, and content having a difficulty level not exceeding the capability value of the user is provided to the outside. The capability filter type content recommendation device according to claim 1. An evaluation option for the user to explicitly give an evaluation to the content is given to at least a part of the content provided to the user, a selection input by the user who has browsed the content is taken in, and the difficulty estimation processing unit An evaluation data capture unit to be reflected in the estimation of difficulty,
The capability filter type content recommendation device according to claim 1, comprising: A dictionary that can be referred to for meaning or reading is assigned to the content with respect to kanji or words having high difficulty in the content, the number of times the user has referred to the dictionary is taken in, and the difficulty by the difficulty determination processing unit An evaluation data capturing unit to be reflected in the degree estimation,
The capability filter type content recommendation device according to claim 1, comprising: The difficulty level estimation processing unit
The initial difficulty level preset for each content in the content information database is updated according to a predetermined criterion based on the data fetched by the evaluation data fetching unit. The capability filter type content recommendation device according to any one of the above. The capability estimation processing unit
The capability value preset for each user in the capability information database is updated according to a predetermined criterion based on the data captured by the evaluation data capturing unit. The capability filter type content recommendation device according to claim 1. A content information database storing content and information indicating the difficulty level of the content; and a user capability information database storing information indicating the capability of the user. A content filtering method for recommending content that is recommended,
A browsing request is received from the outside, content is sequentially acquired from the content information database, and information indicating the difficulty level of the acquired content is compared with information indicating the ability of the user stored in the user capability information database. Providing the acquired content to the outside when browsing is permitted;
Collecting a user's evaluation of the provided content, estimating the difficulty of the content from the collected evaluation, information indicating the user's ability, and browsing history, and storing the content in the content information database;
Estimating a new user ability value from information indicating the degree of difficulty of the estimated content and the user ability, and storing it in the content information database;
A capability filter type content recommendation method comprising: A program used for a capability filter type content recommendation device having a content information database storing content and information indicating the difficulty level of the content, and a user capability information database storing information indicating a user capability ,
A browsing request is received from the outside, content is sequentially acquired from the content information database, and information indicating the difficulty level of the acquired content is compared with information indicating the ability of the user stored in the user capability information database. A process of providing the acquired content to the outside when browsing is permitted;
A process of collecting user evaluation of the provided content, estimating the difficulty of the content from the collected evaluation, information indicating the user's ability, and browsing history, and storing the content in the content information database;
A process of estimating a new user ability value from the estimated content difficulty level and information indicating the user ability, and storing the estimated value in the content information database;
A program that causes a computer to execute.

Images