JP2007164530A - Information recommendation device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information recommendation device allowing a user to select information fit for taste of the user without directly setting a retrieval keyword. <P>SOLUTION: A questionnaire presentation means 330 presents a questionnaire to a user terminal 100. When the user answers the questionnaire and transmits a question thereabout to a processing server 300, a keyword extraction means 350 extracts the keyword from a keyword storage means 340 on the basis of answer contents. A content presentation means 360 searches a content storage means 310 by the extracted keyword to extract a content, calculates a recommendation degree of each the extracted content on the basis of the keyword used for the retrieval of the content and the answer contents about the question corresponding to the keyword, and performs presentation of the content to the user terminal 100 in order of height of the recommendation degree. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a technique for recommending information suitable for a user based on a questionnaire result answered by the user.

  In recent years, with the spread of the Internet, many people are searching for favorite information from a large number of Web sites. In particular, recently, attempts have been made to analyze the browsing history and the like, determine the user's preference, and provide information that the user is likely to like.

  However, in the conventional method, there is a problem that if the user makes an incorrect access or the like, noise is included in the analysis result and accurate analysis cannot be performed.

  In view of the above, it is an object of the present invention to provide an information recommendation device that allows a user to select information that suits the user's preference without directly setting a search keyword. .

  In order to solve the above-described problems, in the present invention, content to be searched is stored so that search is possible by keyword, keyword storage that stores a questionnaire question and the keyword in association with each other, and a user A questionnaire presenting means for presenting the questionnaire questions, and after receiving an answer to the presented questionnaire question, a keyword corresponding to each question is extracted from the keyword storage means based on the content of the received answer A keyword extracting unit that performs a search using the extracted keyword, extracts content from the content storage unit, and uses the keyword used for searching the content and the content of an answer to a question corresponding to the keyword To calculate the recommendation level of each extracted content Provides information recommendation apparatus has a configuration having a content presentation means for performing the presentation of content in descending order of the recommendation level.

  According to the present invention, since information is extracted according to the answer to the user's questionnaire, it is possible for the user to select information that suits the user's preference without directly setting a search keyword. It has the effect of becoming.

(1. Device configuration)
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the drawings. FIG. 1 is a configuration diagram showing an embodiment of an information presentation apparatus according to the present invention. In FIG. 1, 100 is a user terminal, 200 is a network, 300 is a processing server, 310 is content storage means, 320 is a network interface, 330 is questionnaire presentation means, 340 is keyword storage means, 350 is keyword extraction means, and 360 is Content presentation means.

  The user terminal 100 is a terminal device capable of transmitting and receiving data to and from the processing server 300 via the network 200, and can be realized by a general-purpose computer having an Internet connection function. The network 200 is a communication line for transmitting and receiving data between computers, and corresponds to the Internet in this embodiment.

  The processing server 300 has functions as a Web server for transmitting / receiving data to / from the user terminal 100 and a database server for handling a database, and includes a content storage unit 310, a network interface 320, and a questionnaire presentation unit 330. , A keyword storage unit 340, a keyword extraction unit 350, and a content presentation unit 360.

  The content storage unit 310 is a storage unit that stores contents such as documents and images that are intended to be viewed by the user. In the present embodiment, data including an image and a document is stored as content. The network interface 320 is an interface for transmitting / receiving data to / from the user terminal 100 via the network 200. The questionnaire presentation unit 330 has a function of extracting questionnaire screen data from a predetermined storage unit in response to a request from the user terminal 100 and presenting it to the user terminal 100.

  The keyword storage means 340 stores keywords for searching for content in association with questionnaire questions. Here, an example of the keyword for each question stored in the keyword storage unit 340 is shown in FIG. As shown in FIG. 2, in this embodiment, a plurality of keywords are set for each question in the questionnaire. For example, in the question (1), keywords such as “sea” and “beach” are set.

  The keyword extraction unit 350 has a function of extracting a keyword corresponding to each question from the keyword storage unit 340 based on the content of the answer to the presented questionnaire question. The content presentation unit 360 searches the content storage unit 310 using the keyword extracted by the keyword extraction unit 350 to extract the corresponding content, and corresponds to the keyword used for searching the extracted content and the keyword. It has a function of calculating the recommendation level of each extracted content based on the content of the answer to the question and presenting the content in descending order of recommendation level.

(2. Processing operation)
Next, the processing operation of the system shown in FIG. 1 will be described. FIG. 3 is a flowchart showing an outline of the processing operation of the system shown in FIG. First, when the processing terminal 300 is accessed from the user terminal 100, the processing server 300 recognizes the accessing user. In the present embodiment, the user terminal 100 is requested to input a user ID. Note that user recognition is not limited to authentication by inputting a user ID, and any method may be used as long as the user can be specified such as using cookie information of a browser. .

  When the processing server 300 recognizes the user, the questionnaire presenting means 330 presents a questionnaire question to the accessed user (S1). Specifically, the questionnaire question screen data held by the questionnaire presenting means 330 is transmitted to the access source user terminal 100 via the network interface 320. In the user terminal 100, the screen data of the questionnaire question received by the function of the installed browser is displayed on the screen. Here, the questionnaire screen displayed on the user terminal 100 is shown in FIG. In this embodiment, nine questions for selection and one question for entry are prepared as questionnaire questions. As shown in FIG. 4, there are five choices for nine selection questions, and each question is selected from “not important” to “important” in five stages. Therefore, the user selects “1” if he / she does not attach importance, and selects “5” if he / she attaches great importance.

  The user answers the questions on the questionnaire screen as shown in FIG. The state of the questionnaire screen after answering is shown in FIG. In FIG. 5, for the selectable questions (1) to (9), the circles indicated by shading are the selected answer numbers. In addition, the user inputs characters to the entry-type question (10) directly using a keyboard or the like.

  When the user clicks the send button after answering, the user terminal 100 associates the answer content with the question number and sends it to the processing server 300 as questionnaire answer information. Then, the processing server 300 acquires questionnaire response information from the user terminal 100 (S2). Here, the questionnaire response information is shown in FIG. As shown in FIG. 6A, the questionnaire response information is obtained by associating a user's answer on the questionnaire screen shown in FIG. 5 with a question number.

  Subsequently, the keyword extraction unit 350 extracts keywords corresponding to each question from the keyword storage unit 340 based on the acquired questionnaire response information (S3). At this time, since the question numbers (1) to (9) exist in the keyword storage unit 340 as shown in FIG. 2, the keyword extraction unit 350 extracts all corresponding keywords. . Since the question number (10) is an entry type and character information is recorded in the questionnaire response information, the keyword extracting means 350 applies the character information corresponding to the question number (10) in the questionnaire response information. A keyword is generated by performing morphological analysis.

  Subsequently, the content presentation unit 360 searches the content storage unit 310 using the keyword extracted or generated by the keyword extraction unit 350 and extracts the corresponding content. Further, the recommendation level of each extracted content is calculated based on the keyword used for the retrieval of the extracted content and the answer content for the question corresponding to the keyword (S4). Specifically, the recommendation level is calculated by executing processing according to the following [Equation 1].

[Formula 1]
Recommendation level = Σ (number of matched question keywords x importance of question) + number of matched free keywords

  In the above [Formula 1], the “number of matched question keywords” indicates how many keywords exist in the content for each question of the selection formula, and when there are a plurality of the same keywords, The existing number is added. “Question importance” indicates how much the user places importance on the content of the question, and here, an answer number is used. The “number of matched free keywords” indicates how many free keywords generated from the character information answered to the entry type question exist in the content.

  For example, let us consider a case where a content including a document as shown in FIG. 6B is searched using the question-specific keyword shown in FIG. In this case, the content shown in FIG. 6B includes four keywords “Parthenon”, “Greece”, “World Heritage”, and “Tourism” corresponding to the question number (4). The keywords corresponding to the nine questions are not included at all. As shown in FIG. 6A, since the answer number of the question number (4) is “5”, the weight is also “5”, and the number of hit keywords “4” is multiplied by “20”. The other question numbers are calculated in the same manner, and the values for all questions are summed. However, in this example, the keywords of the other questions are not included at all, and are “0”. However, the value for each question other than the question number (4) is “0”, and the score calculated as the recommendation level is “20” after all.

  Similarly, the content presentation unit 360 calculates a recommendation level for all the extracted content. Then, the content is presented to the user terminal 100 (S5). Specifically, first, list screen data in which a predetermined number of all extracted contents are arranged in order of recommendation is created, and this is transmitted to the access source user terminal 100 via the network interface 320. Here, the content list screen displayed on the user terminal 100 is shown in FIG. FIG. 7 shows a case where the top three items with the highest recommendation level are displayed as an example. In the list screen, items with the highest recommendation level are displayed in order from the top. In FIG. 7, “XX resort” has the highest recommendation level, and “Greek tour” has the third highest recommendation level. Is high. In FIG. 7, the shaded portion is a content image.

  A URL link is set in the image portion of each content displayed on the content list screen shown in FIG. Therefore, when the user clicks on the image portion of the content, the information specified by the set URL is accessed and the details of the corresponding content are displayed.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above embodiments, and various modifications can be made. For example, in the above embodiment, the processing server 300 is accessed from the user terminal 100 via the network, and processing is performed on the processing server 300 side. However, the content storage unit 310 and the questionnaire presentation are provided on a stand-alone computer. Functions as the means 330, the keyword storage means 340, the keyword extraction means 350, and the content presentation means 360 may be installed, and a content list may be presented to a display device connected to the computer.

It is a block diagram which shows one Embodiment of the information recommendation apparatus which concerns on this invention. It is a figure which shows an example of the keyword according to question memorize | stored in the keyword memory | storage means 340. FIG. It is a flowchart which shows the outline | summary of the processing operation of the system shown in FIG. It is a figure which shows the questionnaire screen displayed on the user terminal. It is a figure which shows the questionnaire screen of the state which the user answered. It is a figure which shows the questionnaire reply information which matched the user's reply content and the question number, and the document in a content. It is a figure which shows the content list screen displayed on the user terminal.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 ... User terminal 200 ... Network 300 ... Processing server 310 ... Content storage means 320 ... Network interface 330 ... Questionnaire presentation means 340 ... Keyword storage means 350 ... Keyword Extraction means 360... Content presentation means

Claims (3)

Content storage means in which content to be searched is stored so as to be searchable by keyword;
Keyword storage means for storing a questionnaire question and the keyword in association with each other;
Questionnaire presenting means for presenting the questionnaire questions to the user;
After receiving an answer to the question of the presented questionnaire, based on the content of the received answer, a keyword extracting means for extracting a keyword corresponding to each question from the keyword storage means;
Search using the extracted keyword, extract content from the content storage means, and based on the keyword used for the search of the content and the answer content for the question corresponding to the keyword, Content presentation means for calculating a recommendation level of the content and presenting the content in descending order of the recommendation level;
An information recommendation device comprising:   2. The content presentation unit according to claim 1, wherein the content presentation unit calculates, as a recommendation level, a value obtained by multiplying the number of keywords corresponding to the extracted content by using a response number for a questionnaire question as a weight. Information recommendation device. A configuration in which a user terminal that is a terminal device for a user to browse content and a processing server that is a server computer that provides content to the user are connected via a network,
The processing server includes the content storage unit, the keyword storage unit, the questionnaire presentation unit, the keyword extraction unit, and the content presentation unit, and highly recommends the corresponding content to the user terminal. The information recommendation device according to claim 1, wherein the information recommendation device is presented in order.

Images