JP2006338509A - Status storage type content management apparatus and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To search for a content according to the situation without imposing a load on the user by managing the content together with a surrounding situation where the content is created.
A situation storage type content management apparatus 100 stores a category, a feature vector indicating a feature of a surrounding situation, and typical content in association with each other. At the time of content generation, the wireless tag ID is read from the wireless IC tag attached to the surrounding article or person, the feature vector indicating the situation closest to the wireless tag ID is selected, and the category corresponding to the selected feature vector is generated. The associated content is stored in association with each other, and the feature vector is updated with the information of the read RFID tag ID. At the time of content search, a feature vector indicating a situation closest to the wireless tag ID read from surrounding wireless IC tags is selected, and information on typical content corresponding to the selected feature vector is output.
[Selection] Figure 1

Description

  The present invention relates to an indexing technique for searching digital contents such as documents and videos and a technique for classifying digital contents, and in particular, a situation storage type content management apparatus and method using situation information when digital contents are generated. About.

  With the widespread use of portable information devices such as mobile phones and laptop computers, various digital contents (hereinafter simply referred to as “contents”) can be recorded on computers easily and in large quantities. Therefore, a technique relating to indexing and content classification for efficiently searching for a content desired by a user even for a large amount of content has been proposed. For example, in Patent Document 1, regardless of the form of content such as a document, an image, and a sound, each content is expressed by a triplet of “heading”, “content”, and “keyword for search”, thereby Organize and search. The keywords in the triple are set in advance by the user, and if the content is text data, it is extracted from the content by natural language processing. Further, the user can classify the contents into preset categories according to the contents. The user can refer to the content while following the category, and can search for the content by the keyword in the triplet or the date / time when the content was created / updated.

In addition, an indexing technique based on information other than text such as sensor information has been proposed. For example, in Patent Document 2, when an image is recorded in a database, indexing is performed using time information and position information acquired by a global positioning system (GPS) and user physiological information acquired by an electroencephalograph. Since GPS and electroencephalograph information is automatically acquired at the time of image registration, the burden on the user is lower than indexing by keywords. Note that information other than this, for example, information on a person or the like shown in the image, is indexed by direct registration by the user. Since the GPS information is collated with map information registered in advance, it is possible to search for images in prefectures, for example. Furthermore, by analyzing the electroencephalogram information, it is possible to search for an image in which the user is interested.
JP-A-5-61914 JP 2002-342357 A

However, the prior art has the following problems.
When referring to past contents in order to solve a problem that the user is currently facing, the surrounding situation when the contents are recorded may be important. The surrounding conditions here are the time, place, and person who was there. For example, when a salesman negotiates with a customer, the method of proceeding with the negotiation may differ depending on whether the other party is a new customer or an existing customer. In addition, there are cases where the topic to be extracted changes depending on whether the customer meets in a formal place or an informal place. When faced with such a situation, there is a high possibility that a business case related to a customer of the same type as the current customer or a business case conducted in the same place is helpful. However, in Japanese Patent Laid-Open No. 2004-26853, information on the surrounding situation other than the time must be directly registered by the user unless the content is described in text, and this work is expensive for the user. Even in Patent Document 2, information on time and place is automatically registered, but information on characters in the image needs to be input by the user. Is very expensive. Further, in an environment such as a room where GPS cannot be used or a place where map information is not maintained, the user needs to directly register the location information.

  Furthermore, when searching for content, selecting an expression that allows the user to assume the surrounding situation as a keyword is also a very burdensome operation for the user. For example, when content is recorded in the reception room in the company, the meaning of “in-house”, “branch office”, “head office”, etc. due to the difference in synonym / synonym levels such as “reception room”, “reception room”, “customer room”, etc. There is a possibility that the location information is registered in various expressions up to the difference in range. In particular, it is very difficult to use a consistent expression when the user who registered the content is different from the user who uses the content, such as when the content is shared within the organization.

  With regard to content classification, since it is necessary to set categories to be classified in advance in Patent Document 1, when content that cannot be classified into existing categories is to be registered, each time the user considers an appropriate category and makes additional settings. This process is expensive when dealing with a large amount of content.

  The present invention has been made in view of such circumstances, and its purpose is to manage the content together with the surrounding circumstances in which the content is created, so that the user can respond to the situation without imposing a load on the user. It is an object of the present invention to provide a situation storage type content management apparatus and method capable of searching for a content.

  In order to solve the above-described problems, the present invention is a situation storage type content management apparatus that manages created content together with information on the surrounding situation where the content is created, and searches for the content according to the surrounding situation. Generated based on the category identification information of the category for classifying the content and the information recording medium identification information read from the surrounding information recording medium when the content belonging to the category is created. Storage means for associating and storing characteristic information to be shown and content identification information of typical content of the category, situation recording means for reading information recording medium identification information from surrounding information recording media, and reading by the situation recording means Based on the information recording medium identification information, feature information indicating the most similar surrounding situation among the feature information in the storage means The category classification means for selecting the content, the content identification information of the typical content in the storage means is specified by the feature information selected by the category classification means, and the output of the information of the content specified by the content identification information is instructed And a status storage type content management device.

  Further, the present invention is the situation storage type content management apparatus described above, wherein the storage means further stores content identification information and category identification information of a category to which the content belongs in association with each other, and the situation recording means includes: The information recording medium identification information is read from the surrounding information recording medium at the time of content creation, and the category classification means is characterized by the feature information in the storage means based on the information recording medium identification information read by the situation recording means at the time of content creation. Feature information indicating the most similar situation is selected, and the control means stores the storage information according to the feature information selected by the category classification means based on the information recording medium identification information read by the situation recording means at the time of content creation. The category identification information in the means is identified, and the identified category identification information and the content In association with tool identification information written in the storage means, characterized in that.

  Further, the present invention is the above-described situation storage type content management apparatus, wherein the storage unit further stores the content identification information in association with the information recording medium identification information read at the time of content creation, and the category The classifying unit further updates the feature information selected based on the information recording medium identification information by using the information recording medium identification information read by the situation recording unit at the time of content creation, and each content in the storage unit Based on the information recording medium identification information read at the time of creation, the content created in a situation most similar to the updated feature information is selected, and the control means includes the content identification information of the created content, and the situation recording The information is recorded in the storage means in association with the information recording medium identification information read when the content is created, and The feature information in the storage means selected by the category classification means based on the information recording medium identification information read by the status recording means at the time of content creation is rewritten to the updated feature information of the category classification means, and also rewritten. Rewriting the content identification information of the typical content in the storage means corresponding to the feature information to the content identification information of the content selected by the category classification means as being created in a situation most similar to the updated feature information. It is characterized by that.

  Further, the present invention is the situation storage type content management apparatus described above, wherein the category classification means determines that there is no similar feature information based on information recording medium identification information read by the situation recording means at the time of content creation. In the case where the feature information is generated based on the information recording medium identification information, and the control means determines that there is no similar feature information, the content identification information of the created content The category identification information of the new category is associated and written to the storage means, the category identification information of the new category, the content identification information of the created content as typical content, and the feature information generated by the category classification means Are written in the storage means in association with each other.

  Further, the present invention is the above-described situation storage type content management apparatus, wherein the category includes a basic category classified by an object to which the information recording medium is attached and a category to which the object to which the information recording medium is attached belongs. And the storage means stores the identification information of the object category to which the information recording medium is attached in association with the information recording medium identification information, and stores the basic information. The feature information of the category is generated based on information on whether each information recording medium identification information stored in the storage means is read by the situation recording means, and the feature information of the object category is: The identification information of each object category stored in the storage means is the status recording means Is generated based on whether the information in the read information identifying information of the object category corresponding to the recording medium identification information, and wherein the.

  In order to solve the above-described problems, the present invention is used in a situation storage type content management apparatus that manages created content together with information on the surrounding situation in which the content is created and searches for the content according to the surrounding situation. A content storage type content management method that is based on category identification information of a category for classifying content and information recording medium identification information read from surrounding information recording media when content belonging to the category is created And a storage means for storing the characteristic information indicating the characteristics of the surrounding situation and the content identification information of the typical content of the category in association with each other, and reading and reading the information recording medium identification information from the surrounding information recording medium. Based on the information recording medium identification information, the most similar surrounding situation among the characteristic information in the storage means Characteristic information to be displayed is selected, content identification information of typical content in the storage means is specified by the selected feature information, and output of content information specified by the content identification information is instructed. This is a situation storage type content management method.

  Further, the present invention is the above-described situation storage type content management method, wherein the storage unit further stores the content identification information and the category identification information of the category to which the content belongs in association with each other, The information recording medium identification information is read from the information recording medium, and based on the information recording medium identification information read at the time of content creation, the feature information indicating the most similar situation is selected from the characteristic information in the storage unit, and the selected The category identification information of the category in the storage means is specified by the feature information, and the identified category identification information and the content identification information of the created content are associated and written to the storage means.

  Further, the present invention is the above-described situation storage type content management method, wherein the storage means further stores the content identification information in association with the information recording medium identification information read at the time of content creation, and creates the content The feature information selected based on the information recording medium identification information is updated using the information recording medium identification information read at times, and based on the information recording medium identification information read at the time of creating each content in the storage means Selecting the content created in a situation most similar to the updated feature information, associating the content identification information of the created content with the information recording medium identification information read at the time of content creation, and writing to the storage means; The characteristics in the storage means selected based on the information recording medium identification information read at the time of content creation The information is rewritten with the updated feature information, and the content identification information of the typical content in the storage means corresponding to the rewritten feature information is selected as being created in a situation most similar to the updated feature information. It is characterized by rewriting the content identification information of the content.

  In addition, the present invention is the above-described situation storage type content management method, and when it is determined that there is no similar feature information based on the information recording medium identification information read at the time of content creation, the content identification of the created content The information and the category identification information of the new category are associated and written to the storage means, and the category identification information of the new category, the content identification information of the created content as typical content, and the information record read at the time of content creation It is characterized in that it is written in the storage means in association with the feature information generated based on the medium identification information.

  Further, the present invention is the above-described situation storage type content management method, wherein the category includes a basic category classified by an object to which the information recording medium is attached and a category to which the object to which the information recording medium is attached belongs. The semantic category classified by the object category, and the storage means stores the identification information of the object category to which the target to which the information recording medium is attached in association with the information recording medium identification information, and stores the basic information The category feature information is generated based on information on whether or not each information recording medium identification information stored in the storage means has been read from the surrounding information recording medium, and the object category feature information is The identification information of each object category stored in the storage means is a peripheral information recording medium. Is generated based on whether the information is the identification information of the object category corresponding to the information recording medium identification information read from, characterized in that.

  According to the present invention, the situation storage type content management apparatus expresses the surrounding situation when the content is created by the presence or absence of an article or the presence of a person, and the surrounding situation when created is similar Are classified into the same category. At this time, if the newly created content cannot be classified into an existing category, a new category is automatically created and classified. This makes it possible to search for content created in a situation similar to a certain surrounding situation. Furthermore, by identifying typical content that most expresses the characteristics of each category, it is possible to assist the user in understanding the meaning of the category. In addition, it is possible to grasp the surrounding situation in a limited and general manner, and to classify and search the contents according to each way.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing the overall configuration of a situation storage type content management apparatus 100 according to an embodiment of the present invention. In FIG. 1, the situation storage type content management apparatus 100 includes an input / output device 101, a content creator 102, a situation recorder 103, a tag database 104, a category classifier 105, a content database 106, and a controller 107.

  The input / output device 101 inputs various parameters related to content recording and content search, and displays the content search results. The content creator 102 creates content. The situation recorder 103 records the surrounding situation when the content is created. The tag database 104 stores information on wireless tags attached to articles and people in order to record the situation. The category classifier 105 specifies the classification destination of the content according to the surrounding situation when the content is created and the classification destination of the content created in a situation similar to a certain surrounding situation. The content database 106 stores information on content and its classification result. The controller 107 controls input / output and searches a database.

  In the present embodiment, the situation storage type content management apparatus 100 expresses the surrounding situation when the content is recorded by the presence of an article or a person at the place, thereby solving the problem of situation expression selection. Resolve. Furthermore, an active wireless IC tag (hereinafter referred to as “wireless tag”) as an information recording medium is attached to an article or a person, and the situation storage type content management apparatus 100 stores the content in the wireless tag when recording the content. This information is automatically recorded internally as the surrounding situation by receiving the wireless tag ID stored in the information using a receiver (hereinafter referred to as “wireless tag reader”) constituting the situation recorder 103. To do. Note that the wireless tag ID to be used is information recording medium identification information, and is unique to each wireless tag without duplication. Information on the wireless tag is described, for example, in the supervision of Junichi Kishigami, “All of the wireless IC tags for the point-based RFID textbook ubiquitous society”, ASCII, 2005 (Reference 1).

  In the situation storage type content management apparatus 100, the contents similar to the surrounding conditions at a certain point in time are created by classifying contents having similar surrounding conditions when the contents are created into the same category. You can search for content. When classifying content, classification is started from a state where there are no categories. If newly created content cannot be classified into an existing category, a new category is automatically created and classified into that category. . Therefore, it is not necessary for the user to add a new category in consideration. Furthermore, in the process of categorizing the content one after another, it is possible to identify the content that most significantly expresses the characteristics of each category (hereinafter referred to as “typical content” of the category). By referencing, the user can easily understand the feature of the category.

Next, the association between the wireless tag, the article, and the person will be described with reference to FIG.
FIG. 6 is a diagram showing information on the wireless tag used by the situation storage type content management apparatus 100 of the present invention. In the figure, the wireless tag with the wireless tag ID = T000001 is attached to “Mr. X's personal computer” and is associated with the data of the name “Mr. X's personal computer”. Further, “Mr. X's personal computer” is a kind of “personal computer”, and therefore, a category (category ID = OC000001) meaning “personal computer” is associated with a wireless tag of wireless tag ID = T000001. Similarly, when a person to whom a wireless tag is attached is regarded as a member of a group such as an organization, a category representing a group is associated with the person. Hereinafter, such a category representing an article type or group will be referred to as an object category. The wireless tag and the article / individual are associated with each other one-to-one, and the wireless tag and the object category are associated one-to-one. Hereinafter, a category classified by the surrounding situation expressed by the former association is called a basic category, and a category classified by the surrounding situation expressed by the latter association is called a semantic category. Basic categories are classified according to limited surrounding situations composed of specific articles and people, and semantic categories are classified according to semantically generalized surrounding situations composed of articles and groups of the same type. . By switching between these two categories depending on the problem, it is possible to classify and search content that changes the way the surrounding situation is perceived. For example, it is assumed that there are a conference room A and a conference room B, and both have the same type of article. When a user in the conference room A searches for content created in an ambient situation similar to the current ambient situation, using the basic category classification, only the content created in the conference room A is searched. If the semantic category classification is used, in addition to the content created in the conference room A, the content created in the conference room B can be searched.

FIG. 2 is a diagram showing an example of the structure of the tag database 104 of the situation storage type content management apparatus 100 shown in FIG.
In the figure, a tag database 104 holds a tag table that is information that associates a wireless tag ID, a name that identifies an article or an individual, and an object category ID.

FIG. 3 is a diagram showing an example of the structure of the content database 106 of the situation storage type content management apparatus 100 shown in FIG.
In the figure, the content database 106 includes a content table, a content category table, a basic category table, and a semantic category table.
The content table holds information that associates a content ID that specifies content with a wireless tag ID read from the surroundings when the content is generated. The content category table holds information that associates a content ID, a basic category ID that specifies a basic category of a content classification destination, and a semantic category ID that specifies a semantic category of a content classification destination. The basic category table holds information associating basic category IDs, content IDs of typical contents of the basic categories, and feature vector elements w ₁ ,..., W _N expressing features of the basic categories. The semantic category table holds information associating semantic category IDs, content IDs of typical content of semantic categories, and feature vector elements w ₁ ,..., W _N expressing features of the semantic categories. The feature vector is feature information indicating the feature of the surrounding situation, and details will be described later.

Next, content registration will be described with reference to FIG. FIG. 4 is a flowchart showing an outline of the operation in content registration of the situation storage type content management apparatus 100 shown in FIG.
The user inputs parameters related to content classification by the input / output device 101 (STEP 101), and creates necessary content by the content creation device 102 constituted by a video camera or the like (STEP 102). When the content information created by the content creator 102 is output to the controller 107, the controller 107 instructs the situation recorder 103 constituted by a wireless tag reader to record the surrounding situation when the content is created. To do. The status recorder 103 receives radio waves from the wireless tag within a predetermined time set in advance, and outputs the received wireless tag ID to the controller 107 (STEP 103).

  The controller 107 registers the wireless tag ID input by the status recorder 103 and the content ID of the content created by the content creator 102 in the content table in the content database 106. In addition, the controller 107 searches the tag table in the tag database 104 for the object category ID associated with the wireless tag ID (STEP 104), and obtains information related to content classification in the content database 106, the basic category table, and the content category table in the content database 106. Read from the semantic category table and output them to the category classifier 105. The category classifier 105 performs content classification based on the information input by the controller 107, and outputs the content classification category to the controller 107 (STEP 105). The controller 107 searches the content table for information on the RFID tag ID related to the content classified into the category to be classified, and outputs the result to the category classifier 105. The category classifier 105 identifies typical content from the information of the wireless tag ID and outputs the result to the controller 107 (STEP 106). The controller 107 registers the result of content classification and the result of specifying typical content in the content category table, basic category table, and semantic category table (STEP 107).

Next, content search will be described with reference to FIG.
FIG. 5 is a flowchart showing an outline of an operation in content search of the status storage type content management apparatus 100 shown in FIG. The user inputs information on search start conditions such as when and when to start a content search using the input / output device 101 (STEP 201). When the controller 107 instructs the situation recorder 103 to record the surrounding situation when the search start condition is satisfied, the situation recorder 103 receives radio waves from the wireless tag within a predetermined time period. The received wireless tag ID is output to the controller 107 (STEP 202). The controller 107 searches the tag table for the object category ID related to the wireless tag ID input by the status recorder 103 (STEP 203), reads information related to content classification from the basic category table and the semantic category table, and reads them out of the category classifier 105. Output to. The category classifier 105 identifies the basic category and semantic category that are most similar to the surrounding situation recorded by the situation recorder 103 based on the information input by the controller 107, and outputs the result to the controller 107. (STEP 204). The controller 107 searches the content category table for the content classified into the basic category and the semantic category input by the category classifier 105, and further searches the basic category and the semantic category table for typical content for the basic category and the semantic category. When these search results are output to the input / output device 101 (STEP 205), the input / output device 101 displays the search results on the screen and presents them to the user.

Next, the processing of the category classifier 105 will be described in detail.
First, the operation of the category classifier 105 in content registration (FIG. 4) will be described.
The category classifier 105 expresses the surrounding situation by a vector having the wireless tag ID and the object category ID input from the controller 107 as elements. However, for an article or a person for whom no object category is set, the wireless tag ID is used as the object category ID for convenience in order to effectively use the fact that the object category exists. Hereinafter, a vector expressing the surrounding situation by the wireless tag ID is called a basic vector, and a vector expressing the surrounding situation by the object category ID is called a semantic vector. The number of dimensions of the basic vector is the number of wireless tags registered in the tag table, and the wireless tag has a one-to-one correspondence with each element of the vector. Here, when an ID of a certain wireless tag is inputted from the controller 107, the element of the basic vector corresponding to the wireless tag is set to 1, and when not inputted, it is set to 0. Similarly, the number of dimensions of the semantic vector is the number of object categories registered in the tag table, and the object category has a one-to-one correspondence with each element of the vector. Here, when an ID of an object category is inputted from the controller 107, the element of the semantic vector corresponding to the object category is set to 1, and when not inputted, 0 is set. Basic vectors are used for classification by basic categories, and semantic vectors are used for classification by semantic categories. Since the category classification method does not change between the basic category and the semantic category, the basic category will be described below as an example. Categories include GA Carpenter, S. Grossberg, and DB Rosen, “ART 2-A: An adaptive resonance algorithm for rapid category learning and recognition”, Neural Networks, vol.4, pp.493.504, 1991 (reference 2). ) Is used.

The basic vector X ^k of the content k is set as follows.

  Here, N is the number of wireless tags registered in the tag table.

Corresponds to the i-th (i = 1,..., N) wireless tag, and is set to 1 when the ID of the wireless tag is input, and 0 when not input.

In addition, a feature vector W ^j that is an N-dimensional vector expressing the feature of the basic category j is set as follows (j = 1,..., M).

Here, M is the number of basic categories that currently exist. The size of the feature vector W ^j is constant regardless of the basic category j and is 1.

Corresponds to the i-th (i = 1,..., N) wireless tag and indicates the strength of the relative relationship with the wireless tag.

The category classifier 105 updates the basic category feature vector W ^j according to the following equation. In the equations (4) and (5), ρ and η are parameters. The function Ν () is a function for setting the vector size to 1.

First, a conversion vector H ^{k obtained} by converting the magnitude of the basic vector X ^k to 1 is calculated by the equation (1). Then, a numerical value T ^j obtained by adding the product of each element is calculated from the calculated conversion vector H ^k and the feature vector W ^{j of} each basic category j (j = 1,..., N) by the equation (2). , (3) the expression, the maximum value ^{T J} of ^{T j} calculated by equation (2). The feature vector W ^j of each basic category j (j = 1,..., N) is read from the basic category table in the content database 106 by the controller 107 and output to the category classifier 105. When the expression (4) is satisfied in the basic category J where T ^j is the maximum, that is, when the maximum value T ^J calculated by the expression (3) is equal to or larger than the parameter ρ, the content k is classified into the basic category J. , The feature vector ^WJ of the current basic category ^J is updated according to the feature vector WJ ^(new) calculated by the equation (5).

Further, the category classifier 105 outputs the basic category to which the content k is classified as J to the controller 107, and then has already been classified into the basic category J input from the controller 107 (in the following description, Equation (6) is calculated based on the information of the wireless tag ID (assuming that there are L). That is, the controller 107 acquires the content ID of the content already classified in the basic category J from the content category table, acquires information on the wireless tag ID corresponding to the acquired content ID from the content table, and determines the category classification. Output to the device 105. The category classifier 105 generates L basic vectors X ^k (k = 1,..., L) corresponding to each content from the input RFID tag ID information of each L content. Category classifier 105, the L number of basis vectors X ^{k (k} = 1, ..., L) and translation vector H ^k calculated using the equation (1), feature vector W basic categories J updating in the ^{For J} , a numerical value T ^{k obtained} by adding the product of each element is calculated. Further, the maximum value T ^K of T ^k calculated by the equation (6) is obtained from the equation (7). The content K having the basic vector whose T ^k calculated by the equation (6) is the maximum value T ^K is set as a typical content of the basic category J. The basic category of the classification destination of the content k, the update result of W ^j , and the typical content specification result are output to the controller 107. In the controller 107, it rewrites the value element is the updated feature vector W ^J in the basic category table, further, the typical content ID, and updates the content ID of the content K. Further, an entry having information on the generated content ID of the content k and the basic category ID of the basic category J that is the classification destination is inserted into the content category table.

If the expression (4) is not satisfied, a new basic category M + 1 is created with W ^{M + 1} = H ^k , and the content k is classified into the basic category M + 1. In this case, the content k is a typical content of the basic category M + 1, the current value of M is increased by 1, and the result is output to the controller 107. In the controller 107, an entry having information on the basic category ID of the basic category M + 1, the element of the feature vector of the basic category M + 1, and the typical content ID indicating the content ID of the content k is added to the basic category table. . Further, an entry having information on the content ID of the content k and the basic category ID of the basic category M + 1 is inserted into the content category table.

In this way, content can be classified into basic categories having similar basic vectors, and when there are no basic categories that can be classified, the basic categories can be automatically added to classify the categories. Further, as the category classification progresses, W ^j converges to the average vector in the basic vectors of the contents classified into the basic category j, so that the content having the basic vector most similar to W ^j is selected as the representative of the basic category j. Content. Note that the parameter ρ is a parameter related to identification of the basic category, and if this value is large, the number of basic categories increases because the number of cases where the expression (4) is not satisfied increases. In addition, when there are cases where radio waves from radio tags that should be able to be received due to the influence of the radio wave environment may not be received, the effect of the environment can be reduced by gradually updating W ^J by decreasing the value of η. it can.

Next, the operation of the category classifier 105 in content search (FIG. 5) will be described.
The category classifier 105 creates a basic vector X ^k from the information input by the controller 107. Then, according to the equation (2), the transformation vector H ^k calculated from the basic vector X ^{k in the} equation (1) and the feature vector W ^{j of} each basic category j (j = 1,..., N) A numerical value T ^j obtained by adding the products is calculated, and a maximum value T ^J of T ^j calculated by the equation (2) is obtained from the equation (3). Note that the feature vector W ^j for each basic category j (j = 1,..., N) is read from the basic category table by the controller 107 and output to the category classifier 105. When the expression (4) is satisfied in the basic category J where T ^j is the maximum value T ^J , content created in a situation similar to the surrounding situation represented by the input information is displayed for the basic category J. The classified basic category is output to the controller 107. On the other hand, when the expression (4) is not satisfied, it is output to the controller 107 as “no applicable basic category”.

  The above content registration and content search processing is executed using the basic category as the semantic category and the basic vector as the semantic vector, thereby performing content classification and content search by the semantic category. Further, the basic category ID is read as a semantic category ID, and the basic category table is read as a semantic category table.

  As described above, according to the present embodiment, the situation storage type content management apparatus 100 is created by automatically expressing the surrounding situation when the contents are created depending on the presence or absence of articles and the presence of people. Content that is similar to the surrounding environment is classified into the same category. If newly created content cannot be classified into an existing category, a new category is automatically created and classified. As a result, the situation storage type content management apparatus 100 can search for contents created in a situation similar to a certain surrounding situation. In addition, it is possible to grasp the surrounding situation in a limited and general manner, and to classify and search the contents according to each way. Furthermore, by identifying typical content that most expresses the characteristics of each category, it is possible to assist the user in understanding the meaning of the category.

  The above-described situation storage type content management apparatus 100 has a computer system therein. The operation process of the situation storage type content management apparatus 100 described above is stored in a computer-readable recording medium in the form of a program, and the computer system reads and executes this program, whereby the above-described processing is performed. Is called. The computer system here includes an OS and hardware such as peripheral devices.

  In addition to ROM, “computer-readable recording medium” refers to a portable medium such as a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, or a storage device such as a hard disk built in a computer system. That means. Furthermore, the “computer-readable recording medium” refers to a volatile memory (RAM) inside a computer system that becomes a client or a system when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.

Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” refers to a storage device such as a flexible medium, a magneto-optical disk, a portable medium such as a ROM and a CD-ROM, and a hard disk incorporated in a computer system. Furthermore, the “computer-readable recording medium” dynamically holds a program for a short time like a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line. In this case, a volatile memory in a computer system serving as a server or a client in that case, and a program that holds a program for a certain period of time are also included. The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system.

  The situation storage type content management apparatus according to the embodiment of the present invention has been specifically described above. However, the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention. Needless to say. For example, in this embodiment, an active wireless tag that emits radio waves is assumed as a wireless tag. However, it can be replaced with a passive wireless tag that returns an ID in response to a radio wave emitted by a wireless tag reader.

It is a block diagram which shows the whole structure of the situation storage type content management apparatus by one embodiment of this invention. It is a figure which shows an example of the structure of the tag database by the embodiment. It is a figure which shows an example of the structure of the content database by the embodiment. It is a flowchart which shows the outline | summary of the operation | movement in the content registration by the embodiment. It is a flowchart which shows the outline | summary of the operation | movement in the content search by the embodiment. It is a figure which shows the information of the radio | wireless tag which the situation storage type content management apparatus by the embodiment uses.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 ... Input / output device 102 ... Content creation device 103 ... Situation recorder (situation recording means)
104 ... Tag database (storage means)
105 ... Category classifier (category classification means)
106 ... Content database (storage means)
107: Controller (control means)

Claims (10)

A situation storage type content management apparatus that manages the created content together with information on the surrounding situation in which the content is created, and searches for the content according to the surrounding situation,
A feature that is generated based on category identification information of a category for classifying content and information recording medium identification information read from the surrounding information recording medium when the content belonging to the category is created, and that shows the characteristics of the surrounding situation Storage means for storing information in association with content identification information of typical content of the category;
Status recording means for reading information recording medium identification information from surrounding information recording media;
Based on the information recording medium identification information read by the situation recording means, category classification means for selecting feature information indicating the most similar situation among the feature information in the storage means;
Control means for specifying content identification information of typical content in the storage means based on the feature information selected by the category classification means, and instructing output of content information specified by the content identification information;
A situation storage type content management apparatus comprising: The storage means further stores content identification information in association with category identification information of a category to which the content belongs,
The status recording means reads information recording medium identification information from a surrounding information recording medium at the time of content creation,
The category classification unit selects feature information indicating the most similar situation among the feature information in the storage unit based on the information recording medium identification information read by the situation recording unit when creating the content,
The control means specifies the category identification information in the storage means based on the feature information selected by the category classification means based on the information recording medium identification information read by the situation recording means at the time of content creation, and specifies the identified category identification Writing the information and the content identification information of the created content in the storage means in association with each other.
The situation storage type content management apparatus according to claim 1. The storage means further stores content identification information in association with information recording medium identification information read at the time of content creation,
The category classification unit further updates the feature information selected based on the information recording medium identification information using the information recording medium identification information read by the situation recording unit at the time of content creation, and stores the information in the storage unit. Based on the information recording medium identification information read at the time of each content creation, select the content created in the situation most similar to the updated feature information,
The control means associates the content identification information of the created content with the information recording medium identification information read by the situation recording means at the time of content creation and writes it in the storage means, and further, the situation recording means at the time of content creation The feature information in the storage means selected by the category classification means based on the read information recording medium identification information is rewritten to the updated feature information of the category classification means, and the storage corresponding to the rewritten feature information Rewriting the content identification information of the typical content in the means to the content identification information of the content selected by the category classification means as being created in a situation most similar to the updated feature information,
The situation storage type content management apparatus according to claim 2. The category classification unit generates feature information based on the information recording medium identification information when the situation recording unit determines that there is no similar characteristic information based on the information recording medium identification information read at the time of content creation. ,
When the control unit determines that there is no similar feature information, the control unit writes the content identification information of the created content and the category identification information of the new category in association with each other and writes the new category information to the storage unit. The category identification information of the category, the content identification information of the created content as typical content, and the feature information generated by the category classification means are associated and written to the storage means.
The situation storage type content management apparatus according to claim 2 or claim 3, wherein The category is a semantic category classified by a basic category classified by a target to which the information recording medium is attached and an object category to which a target to which the information recording medium is attached belongs.
The storage means stores identification information of an object category to which a target to which the information recording medium is attached belongs in association with the information recording medium identification information,
The feature information of the basic category is generated based on information on whether or not each information recording medium identification information stored in the storage unit is read by the situation recording unit,
Whether the object category feature information is that the identification information of each object category stored in the storage means is the identification information of the object category corresponding to the information recording medium identification information read by the status recording means Generated based on the information of
The situation storage type content management apparatus according to any one of claims 1 to 4, wherein A situation storage type content management method used for a situation storage type content management apparatus that manages created content together with information on the surrounding situation where the content is created and searches for the content according to the surrounding situation.
A feature that is generated based on category identification information of a category for classifying content and information recording medium identification information read from the surrounding information recording medium when the content belonging to the category is created, and that shows the characteristics of the surrounding situation Storage means for storing information in association with content identification information of typical content of the category,
Read information recording medium identification information from the surrounding information recording medium,
Based on the read information recording medium identification information, select the feature information indicating the most similar surrounding situation among the feature information in the storage means,
Specifying the content identification information of the typical content in the storage means by the selected feature information, and instructing the output of the information of the content specified by the content identification information;
A situation storage type content management method. The storage means further stores content identification information in association with category identification information of a category to which the content belongs,
Read the information recording medium identification information from the surrounding information recording medium when creating the content,
Based on the information recording medium identification information read at the time of content creation, select feature information indicating the most similar situation among the feature information in the storage means,
The category identification information of the category in the storage means is specified by the selected feature information, and the specified category identification information and the content identification information of the created content are associated and written to the storage means.
The situation storage type content management method according to claim 6. The storage means further stores content identification information in association with information recording medium identification information read at the time of content creation,
Using the information recording medium identification information read at the time of content creation, the feature information selected based on the information recording medium identification information is updated, and the information recording medium identification information read at the time of content creation in the storage means is updated. Based on the content that is most similar to the updated feature information, select the content created,
The storage means selected based on the information recording medium identification information read at the time of content creation, in association with the content identification information of the created content and the information recording medium identification information read at the time of content creation. The content identification information of the typical content in the storage means corresponding to the rewritten feature information is created in a situation most similar to the updated feature information. Rewrite to the content identification information of the content selected as
The situation storage type content management method according to claim 7. When it is determined that there is no similar feature information based on the information recording medium identification information read at the time of content creation, the content identification information of the created content and the category identification information of the new category are associated and written to the storage means The storage means associates the category identification information of the new category, the content identification information of the created content as typical content, and the feature information generated based on the information recording medium identification information read at the time of content creation. Write,
9. The situation storage type content management method according to claim 7 or 8, wherein: The category is a semantic category classified by a basic category classified by a target to which the information recording medium is attached and an object category to which a target to which the information recording medium is attached belongs.
The storage means stores identification information of an object category to which a target to which the information recording medium is attached belongs in association with the information recording medium identification information;
The feature information of the basic category is generated based on information on whether or not each information recording medium identification information stored in the storage means has been read from the surrounding information recording medium,
The characteristic information of the object category includes identification information of each object category stored in the storage means in identification information of the object category corresponding to information recording medium identification information read from the surrounding information recording medium. Generated based on whether or not,
The situation storage type content management method according to any one of claims 6 to 9, wherein

Images