JP2008171244A - Content retrieval device and method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To properly retrieve a content such as an image with sensitive expressions. <P>SOLUTION: A first database DB1 in which a plurality of types of color pattern and a text having meaning relevant to each color pattern are associated with each other and a second database DB2 in which a plurality of types of content and at least one text having meaning relevant to each content are associated with each other are prepared. When the input of a color pattern for retrieval is accepted by an input part 2, a retrieval part 3 retrieves the text with which the color pattern for retrieval is associated by referring to the first database DB1, and retrieves the content associated with the retrieved text by referring to the second database DB2. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a content search apparatus and method for searching for content such as images, and a program for causing a computer to execute the content search method.

  Various systems for searching contents such as images and moving images from the Internet have been proposed. For example, the URL of an image is extracted from an HTML document collected from a website on the Internet, the image is acquired by referring to the URL, the importance of the acquired image and the HTML document using the image is calculated, and the HTML document The image database is constructed by extracting the image features and image tag information necessary for image search from the image file as search keywords, searching the image database for images that match the accepted search conditions, A method for displaying images in order of importance of images has been proposed (see Patent Document 1).

  In addition, a keyword is searched for an image by determining a search keyword from the input search word, an emotion pattern corresponding to the search word is acquired, an image is searched using the feature amount of the emotion pattern, and two search results are obtained. There has been proposed a technique for accurately searching for a target image by integrating (see Patent Document 2).

According to these methods, a desired image can be searched with high accuracy by inputting a search word.
JP 2004-220267 A JP 2000-112956 A

  However, a user who performs a search may not only search for a search term that expresses an image of a desired image, but also search using a search term that has an ambiguous expression. is there. In the technique described in Patent Document 1, a keyword representing an image held by a user is not necessarily associated with an image required by the user. For this reason, when an emotionally ambiguous expression corresponding to an image included in an image required by the user is used as a search word, an image that is meaningless to the user may be searched.

  Also, in the technique described in Patent Document 2, the information that the user thinks is not necessarily associated as a keyword with the image that the user needs, and similarly, an emotionally ambiguous expression is used as a search term. When used, there is a risk that an image that is meaningless to the user will be retrieved.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to make it possible to appropriately search content such as an image with a sensitive expression.

A content search apparatus according to the present invention includes a first database in which a plurality of types of color patterns and texts having meanings related to the color patterns are associated with each other,
A second database in which a plurality of types of content are associated with at least one text having a meaning associated with each content;
An input means for receiving an input of a search color pattern;
A search for searching for text associated with the search color pattern with reference to the first database and searching for content associated with the searched text with reference to the second database. Means.

  Thus, when a desired search color pattern is input, text having a meaning related to the search color pattern is searched, and content associated with the searched text is searched. For this reason, since the content corresponding to the color pattern which is an emotional expression is searched, it is possible to search for the content reflecting the user's sensitivity.

  The “color pattern” may be a single color or a combination of a plurality of colors.

  As "content", if it can be associated with text having a meaning related to a color pattern, for example, not only still images, moving images, music, books, clothing, and paintings, but also the appearance and design of products. Etc. can also be set as search targets as contents.

  In the content search device according to the present invention, a usage history database in which a score based on the frequency of use of the searched content is associated with text having a meaning related to the plurality of types of color patterns and the plurality of types of content. It may be further provided with history management means for managing as follows.

  Thereby, it is possible to know content with high usage frequency by referring to the usage history database. Further, since the usage history can be accumulated, the past usage history can be reflected in the search result.

  “Use” is to use the content of the search result. Specifically, in the case of still images, moving images, and music, these are played back or downloaded. In the case of books, clothing, paintings, and merchandise, these contents are often searched in online shopping. For this reason, in the case of books, clothing, paintings, and merchandise, “use” may be “use” of content, such as purchase of content as a search result, reservation for purchase, and movement to a shopping basket for purchase.

  In the content search device according to the present invention, the search means may be means for reflecting the score in the search result of the content with reference to the usage history database.

  Thereby, it is possible to obtain a search result reflecting the content usage frequency.

  In the content search device according to the present invention, the history management means may be means for managing the usage history database in units of users.

  Thus, by referring to the usage history database, it is possible to know content with high usage frequency for each user. Further, since the usage history can be accumulated, the past usage history can be reflected in the search result for each user.

  Also, in this case, if there is a text whose score is unknown in the usage history database for the specific user, the history management means refers to the scores of other users other than the specific user to determine the unknown It is good also as a means to estimate the score of a text.

  As a result, when there is a text whose score is unknown in the usage history database for a specific user, the score of the unknown text is estimated by referring to the scores of other users other than the specific user. Can be reflected in search results.

  In the content search apparatus according to the present invention, the search means specifies a user who has input the search color pattern, refers to the use history database of the specified user, and uses the use history as the content history. It is good also as a means to reflect in a search result.

  Thereby, it is possible to obtain a search result reflecting the content usage frequency for a specific user.

In the content search device according to the present invention, the content search apparatus further includes an analysis unit that analyzes a synonym of the searched text.
The search means may be means for searching the content corresponding to the searched text and the text represented by the synonym.

  Thereby, since it is possible to search for content using a wider range of text, it is possible to more easily obtain a search result suitable for the user's intention.

  “Synonyms” include synonyms and related terms of the retrieved text. Synonyms include foreign words, abbreviations, common names, idiomatic expressions, obsolete expressions, fluctuations in notation with kanji and kana, and fluctuations in notation with sending kana.

  The “related word” includes a broad word having a broader meaning than the searched text and a narrow word having a narrow meaning.

  Moreover, in the content search device according to the present invention, the first database is formed by associating a color pattern with text in a first language, and the second database is different from the first language. In the case where the text in the language and the content are associated with each other, a translation means for translating the retrieved text in the first language into the text in the second language may be further provided.

  Thereby, even if the language of the text registered in the first database is different from the language of the text registered in the second database, the content can be searched.

The content search method according to the present invention accepts input of a search color pattern,
With reference to a first database in which a plurality of types of color patterns and texts having meanings related to the respective color patterns are associated with each other, a search is performed for texts associated with the search color patterns. Searching for a content associated with the retrieved text by referring to a second database that associates the content with at least one text having a meaning associated with each content It is.

  The content search method according to the present invention may be provided as a program for causing a computer to execute the content search method.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing the configuration of the content search apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the content search apparatus 1 according to the first embodiment includes an input unit 2 that receives an input of a search color pattern, a first database DB1, a second database DB2, and a search color pattern. Are provided with a search unit 3 for searching for content corresponding to the above, a content storage unit 4 for storing content, and an output unit 5 for outputting a search result of content.

  The input unit 2 is connected to a network such as the Internet, for example, receives a search color pattern transmitted by a user from a terminal device connected to the network, and inputs the search color pattern to the device 1. The search color pattern is input using a search color pattern input screen as shown in FIG. Specifically, the search color pattern input screen 20 shown in FIG. 2 is displayed on the terminal device, and one or more desired colors are designated from a color palette 22 composed of a plurality of color samples 21 on the input screen 20. Then, the search color pattern may be input by clicking the send button 23.

  Alternatively, the user may click on the color creation button 24 to specify the desired hue, saturation, and brightness of the search color pattern with the cursor 25, and an input box 26 for inputting each RGB value. May be designated by inputting the values of RGB colors of a desired color. The color designated by the cursor 25 and the color designated by the input box are displayed in the designated color area 27 by clicking the color creation button 24 after the color is designated. After designating one or more colors, the user clicks the send button 23 to transmit the color pattern from the terminal device to the content search device 1.

  The search unit 3 searches the content corresponding to the search color pattern input by the user with reference to the first and second databases DB1 and DB2.

  The first database DB1 registers a plurality of types of color patterns and texts having meanings related to the color patterns in association with each other. Here, as shown in FIG. 3, the first database DB1 takes the relationship of WARM-COOL as the color impression on the horizontal axis and the relationship of SOFT-HARD on the vertical axis, which is cool, cold, hot, As shown in FIG. 4, a color map that defines the relationship between warm, pop, and plain text and the color pattern impressed by the text is created. Create by associating. In FIG. 3, a color pattern related to the meaning of text such as cool or cold is indicated by “□”.

  In the first database DB1 shown in FIG. 4, “cool 1” is included in the two color patterns (R, G, B) = (0, 0, 255) and (128, 128, 255). , (R, G, B) = (64, 64, 200), (32, 200, 255), the two color patterns are “cool 2”, but (R, G, B) = (32, 255). , 255) and (64, 255, 645), “cool 3” is registered in the two color patterns. Note that the color pattern registered in association with the text is not limited to two colors but may be one color or three or more colors.

  The second database DB2 is a database in which a plurality of types of contents stored in the content storage unit 4 and texts having meanings related to the contents are associated and registered. In the first embodiment, an image is used as the content. In the present embodiment, a survey is performed as to what kind of image the content has, statistics are taken on the text representing the image obtained by the survey, and the text representing the image that more people have for the content. Is a text having a meaning related to the content. Note that not only one text but also a plurality of statistically higher texts may be associated with the content.

  FIG. 5 is a diagram illustrating an example of the second database DB2. As shown in FIG. 5, in the second database DB2, image IDs of a plurality of images and texts having meanings related to contents are registered. For example, the image 001 includes text “cool”, “cool”, “carbonic acid”, “bubble”, “liquid”, and “green”, and the image 002 includes “cool”, “wind”, “cloud”, The texts “Sky”, “Various” and “High” are registered.

  Here, the contents are not limited to images, and moving images, music, books, clothing, paintings, and appearances and designs of products can be used as contents. For these contents, text having a meaning related to the contents may be determined similarly to the image.

  The search unit 3 searches the text associated with the search color pattern input to the input unit 2 with reference to the first database DB1. Specifically, when the color pattern for search and the color pattern associated with the text completely match, the text associated with the completely matched color pattern is included in the search result. For example, when the search color pattern is a color pattern of two colors (R, G, B) = (0, 0, 255), (128, 128, 255), the text “cool” is searched. Include in results.

  If the search color pattern and the color pattern associated with the text do not match, the search color pattern and all the color patterns registered in the first database DB1 are mapped on the RGB color space, The similarity between the search color pattern and the registered color pattern in the RGB color space is calculated, and the text associated with the color pattern having the highest similarity is included in the search result. For example, the distance in the RGB color space between the search color pattern and the registered color pattern is calculated, and the text associated with the color pattern with the shortest distance is included in the search result.

  Specifically, when the color pattern for search and the color pattern for distance determination are composed of two colors, as shown in FIG. 6, the color patterns for search P1 and P2 are mapped on the RGB color space to determine the distance. The color patterns C1 and C2 to be subjected to the above are mapped. Then, the distance L1 between the line segment P1P2 and the line segment C1C2 is obtained for all the color patterns registered in the first database DB1, and the text associated with the color pattern with the shortest distance L1 is included in the search result. .

  It should be noted that not only the color pattern with the shortest distance L1 but also all color patterns are sorted in the order of the shortest distance L1, and the texts associated with the upper predetermined number (for example, three) of color patterns are included in the search result. It may be.

  Here, when the color pattern is composed of two colors, the distance L1 between the line segments described above may be calculated as the similarity, but the similarity is determined by the number of colors of the search color pattern and the color pattern that is the target of the distance determination. Depending on the number of colors, not only the distance between line segments described above, but also the distance between points, the distance between points and lines, the distance between points and surfaces, the distance between points and solids, the distance between lines and surfaces The distance between the line segment and the solid, the distance between the surface and the surface, and the distance between the surface and the solid are different. Moreover, you may make it obtain | require not only the thing based on the distance mentioned above but the degree to which a line segment, a surface, and a solid overlap as a similarity.

  Further, the search unit 3 refers to the second database DB2, searches for content associated with the searched text, and outputs a search result. The content search is performed for all searched texts. When three texts are searched, the content is searched for each of the three texts.

  The output unit 5 is connected to a network such as the Internet, for example, and transmits the search result to the terminal device that has transmitted the search color pattern to the device 1.

  Next, processing performed in the first embodiment will be described. FIG. 7 is a flowchart showing processing performed in the first embodiment. The input unit 2 monitors whether or not the search color pattern has been input (step ST1). When the input of the search color pattern is accepted, the search unit 3 refers to the first database DB1. A text having a meaning related to the search color pattern is searched (step ST2).

  Next, the search unit 3 searches the content associated with the searched text with reference to the second database DB2 (step ST3). And the output part 5 outputs a search result (step ST4), and complete | finishes a process.

  As described above, in the first embodiment, when the search color pattern is input, the text having the meaning related to the search color pattern is searched, and the content associated with the searched text is further searched. It is what I did. For this reason, since the content corresponding to the color pattern which is an emotional expression is searched, it is possible to search for the content reflecting the user's sensitivity.

  Next, a second embodiment of the present invention will be described. FIG. 8 is a schematic block diagram showing the configuration of the content search apparatus according to the second embodiment. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted here. The content search apparatus 1A according to the second embodiment includes a first usage history database DB3 and a plurality of types, in which scores based on usage frequencies of searched contents are associated with texts having meanings related to a plurality of types of color patterns. The second embodiment is different from the first embodiment in that it includes a history management unit 6 that is managed as a second usage history database DB4 associated with the content.

  FIG. 9 is a diagram showing the first usage history database DB3. As shown in FIG. 9, in the first usage history database DB3, a plurality of types of text and usage frequency scores similar to those registered in the first database DB1 are associated. Each time the searched content is used, the history management unit 6 updates the first usage history database DB3 by adding one score of the frequency of the text associated with the content. For example, when the image 001 in the second database DB2 is used, the text searched for the image 001 is “cool”, and the text searched from the search color pattern at that time is “cool 1”. The history management unit 6 adds a score of “cool 1”.

  FIG. 10 is a diagram showing the second usage history database DB4. As shown in FIG. 10, in the usage history database DB4, a plurality of types of content and usage frequency scores similar to those registered in the second database DB2 are associated with each text used for content search. ing. Each time the searched content is used, the history management unit 6 adds the score of the usage frequency corresponding to the text searched for the used content registered in the usage history database DB4 to the second. The usage history database DB4 is updated. For example, when the image 001 is used and the text searched for the image 001 is “cool”, the history management unit 6 adds a “cool” score.

  In the present embodiment, since the content is an image, the content is used to play back or download the content included in the search result.

  Here, in the second embodiment, the search is performed in the same manner as in the first embodiment, but the search unit 3 reflects the use history in the search result. For example, when a plurality of images are searched, the images are displayed side by side in descending order of the score in the search result. Specifically, when the images 001, 002, 003 registered in the second usage history database DB4 are searched and the text used for the search is “cool”, the score of each image is 13, Therefore, the search results are displayed by arranging the images in the order of images 002, 001, 003. In addition, referring to the first usage history database DB3, the search results may be displayed by arranging the images in the descending order of the scores for the text searched for the content searched from the search color pattern.

  Next, processing performed in the second embodiment will be described. FIG. 11 is a flowchart showing processing performed in the second embodiment. The input unit 2 monitors whether or not the search color pattern has been input (step ST11). When the input of the search color pattern is accepted, the search unit 3 refers to the first database DB1. A text having a meaning related to the search color pattern is searched (step ST12).

  Next, the search unit 3 searches the content associated with the searched text by referring to the second database DB2 (step ST13), and further by referring to the second usage history database DB4, The content display order is determined (step ST14). Then, the output unit 5 outputs the search result (step ST15).

  Subsequently, the history management unit 6 starts monitoring whether or not the content in the search result is used (step ST16). When the content is used, the first usage history database DB3 and the first usage history database about the used content are used. 2 usage history database DB4 is updated (step ST17), and the process ends.

  As a result, in the second embodiment, by referring to the second usage history database DB4, the text obtained from the content having a high usage frequency by referring to the first usage history database DB3 is used. You can know the high content in the text unit searched. In addition, since the content usage history can be accumulated, the past usage history can be reflected in the search result.

  Next, a third embodiment of the present invention will be described. FIG. 12 is a schematic block diagram showing the configuration of the content search apparatus according to the third embodiment. Note that in the third embodiment, the same reference numerals are assigned to the same configurations as those in the first embodiment, and detailed description thereof is omitted here. The content search device 1B according to the third embodiment includes a user database DB5 in which users who use the content search device 1B are registered, and the history management unit 6 uses a first usage history database DB3 and a second usage history database DB4. Is different from the second embodiment in that it is managed in units of users.

  FIG. 13 shows the user database DB5. As shown in FIG. 13, the user database DB5 used in the third embodiment has registered therein a user ID, an account name, a mail address, and a name. The input unit 2 refers to the user database DB5 and authenticates a user who performs a search.

  The history management unit 6 manages the first usage history database DB3 and the second usage history database DB4 for each user. Specifically, the first usage history database DB3 and the second usage history database DB4 are managed in association with user IDs, respectively.

  Next, processing performed in the third embodiment will be described. FIG. 14 is a flowchart showing processing performed in the third embodiment. The input unit 2 first authenticates a user who performs a search (step ST21), and then starts monitoring whether or not a search color pattern has been input (step ST22). When the input of the search color pattern is accepted, the search unit 3 refers to the first database DB1 and searches for text having a meaning related to the search color pattern (step ST23).

  Next, the search unit 3 searches the content associated with the searched text with reference to the second database DB2 (step ST24), and further refers to the second usage history database DB4 with reference to the search result. The content display order is determined (step ST25). Then, the output unit 5 outputs the search result (step ST26).

  Subsequently, the history management unit 6 starts monitoring whether or not the content in the search result is used (step ST27). When the content is used, the first of the user who has performed the search for the used content is displayed. The usage history database DB3 and the second usage history database DB4 are updated (step ST28), and the process ends.

  Thus, in the third embodiment, by referring to the first usage history database DB3, by referring to the second usage history database DB4, the text that has obtained the content with high usage frequency for each user is referred to. It is possible to know the frequently used contents for each user by the searched text unit. Further, since the content usage history can be stored for each user, the past usage history can be reflected in the search result for each user.

  In the third embodiment, the first usage history database DB3 and the second usage history database DB4 are managed for each user, but there is a text whose score is unknown in the first usage history database DB3. There is a case. In this case, an unknown score may be estimated using a collaborative filtering technique. Hereinafter, estimation of an unknown score using a collaborative filtering method will be described.

  The collaborative filtering technique is a technique for recording user preferences in the form of past behavior and estimating user preferences based on preference information of users who are taking similar behavior to the user. .

  When the first usage history database DB3 is prepared for each user, a correspondence table between user IDs and text scores can be generated. FIG. 15 is a diagram showing a correspondence table between user IDs and text scores. In the correspondence table shown in FIG. 15, it is assumed that the score of “cold 3” is unknown for the user with the user ID 10001 (hereinafter referred to as user 10001). In this case, the history management unit 6 extracts a user whose score tendency is similar to that of the user 10001 for the text other than “cold 3” from the correspondence table. Looking at the correspondence table shown in FIG. 15, the user 10002 is extracted because the user 10002 has a similar score tendency to the user 10001. Then, the score of “cold 3” for the user 10002 is estimated as the score of “cold 3” for the user 10001.

  Thereby, an unknown score can be estimated and the utilization history of a specific user can be reflected in a search result.

  Next, a fourth embodiment of the present invention will be described. FIG. 16 is a schematic block diagram showing the configuration of the content search apparatus according to the fourth embodiment. Note that in the fourth embodiment, the same reference numerals are assigned to the same configurations as those in the first embodiment, and detailed description thereof is omitted here. The content search apparatus 1C according to the fourth embodiment includes an analysis unit 7 that analyzes synonyms of text searched by the search unit 3 from the search color pattern, and a synonym dictionary database DB6. The text searched by the search unit 3 and its synonyms The difference from the first embodiment is that content corresponding to the text represented by is searched. The fourth embodiment can be similarly applied to the second and third embodiments.

  Here, “synonyms” include synonyms and related terms of the searched text. Synonyms include foreign words, abbreviations, common names, idiomatic expressions, obsolete expressions, fluctuations in notation with kanji and kana, and fluctuations in notation with sending kana.

  The “related word” includes a broad word having a broader meaning than the searched text and a narrow word having a narrow meaning. For example, “emotion” and “emotion” are the broader terms of “impression”, and “impressive” and “ecstasy” are the narrower terms. Further, “emotion” and “emotion” are broadly defined as “fear”, and “awe”, “fright”, “depression”, and the like as narrower terms.

  Synonyms for various texts are registered in the synonym dictionary database DB6, and the analysis unit 7 refers to the synonym dictionary database DB6 and analyzes synonyms of the searched text.

  Next, processing performed in the fourth embodiment will be described. FIG. 17 is a flowchart showing processing performed in the fourth embodiment. The input unit 2 monitors whether or not a search color pattern has been input (step ST31). When the input of the search color pattern is accepted, the search unit 3 refers to the first database DB1. A text having a meaning related to the search color pattern is searched (step ST32). Next, the analysis unit 7 refers to the synonym dictionary database DB6 and analyzes synonyms of the retrieved text (step ST33).

  Next, the search unit 3 refers to the second database DB2 and searches for the content associated with the searched text and the text of the synonym thereof (step ST34). And the output part 5 outputs a search result (step ST35), and complete | finishes a process.

  As a result, in the fourth embodiment, a search can be performed using a wider range of text, so that a search result suitable for the user's intention can be obtained more easily.

  In the fourth embodiment, the content search device 1C includes the synonym dictionary database DB6. However, the synonym dictionary database DB6 is stored in a server provided elsewhere connected to the content search device 1C via a network. Alternatively, synonyms may be analyzed with reference to the synonym dictionary database DB6 via a network.

  Next, a fifth embodiment of the present invention will be described. FIG. 18 is a schematic block diagram showing the configuration of the content search apparatus according to the fifth embodiment. Note that in the fifth embodiment, the same reference numerals are assigned to the same configurations as those in the first embodiment, and detailed description thereof is omitted here. In the content search device 1D according to the fifth embodiment, the first database DB1 has color patterns associated with English text, and the second database DB2 has Japanese text associated with content. In this case, the second embodiment is different from the first embodiment in that a translation section 8 that translates English text searched by the search section 3 into Japanese and a translation dictionary database DB7 are provided. The fifth embodiment can be similarly applied to the second to fourth embodiments.

  Next, processing performed in the fifth embodiment will be described. FIG. 19 is a flowchart showing processing performed in the fifth embodiment. The input unit 2 monitors whether or not the search color pattern has been input (step ST41). When the input of the search color pattern is received, the search unit 3 refers to the first database DB1. An English text having a meaning related to the search color pattern is searched (step ST42). Next, the translation unit 8 translates the retrieved English text into Japanese (step ST43).

  Next, the search unit 3 searches the content associated with the translated text by referring to the second database DB2 (step ST44). And the output part 5 outputs a search result (step ST45), and complete | finishes a process.

  As a result, in the fifth embodiment, even if the text language registered in the first database DB1 and the text language registered in the second database DB2 are different, Search can be performed.

  In the fifth embodiment, only one translation dictionary database DB7 for translating English into Japanese is provided. However, a plurality of translation dictionary databases for translating various languages into Japanese are provided. If prepared, it is possible to search using not only English but also various languages.

  Although the apparatus according to the embodiment of the present invention has been described above, the computer functions as a unit corresponding to the input unit 2, the search unit 3, the output unit 5, the history management unit 6, the analysis unit 7, and the translation unit 8. In addition, a program that causes the processing shown in FIGS. 7, 11, 14, 17, and 19 to be performed is also one embodiment of the present invention. A computer-readable recording medium in which such a program is recorded is also one embodiment of the present invention.

1 is a schematic block diagram showing the configuration of a content search device according to a first embodiment of the present invention. The figure which shows the input screen of the color pattern for search Diagram showing color map Diagram showing the first database Diagram showing the second database The figure for demonstrating calculation of the distance of the color pattern in RGB color space The flowchart which shows the process performed in 1st Embodiment. Schematic block diagram showing the configuration of a content search apparatus according to the second embodiment The figure which shows the 1st use history database The figure which shows the 2nd use history database The flowchart which shows the process performed in 2nd Embodiment. Schematic block diagram showing the configuration of a content search apparatus according to the third embodiment Diagram showing user database The flowchart which shows the process performed in 3rd Embodiment The figure which shows the conversion table of user ID and the score of a text Schematic block diagram showing a configuration of a content search apparatus according to the fourth embodiment The flowchart which shows the process performed in 4th Embodiment Schematic block diagram showing a configuration of a content search apparatus according to a fifth embodiment The flowchart which shows the process performed in 5th Embodiment

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Content search apparatus 2 Input part 3 Search part 4 Content storage part 5 Output part 6 History management part 7 Analysis part 8 Translation part DB1 1st database DB2 2nd database DB3 1st usage history database DB4 2nd usage history Database DB5 User database DB6 Synonym dictionary database DB7 Translation dictionary database

Claims (10)

A first database in which a plurality of types of color patterns are associated with text having meanings associated with the color patterns;
A second database in which a plurality of types of content are associated with at least one text having a meaning associated with each content;
An input means for receiving an input of a search color pattern;
A search for searching for text associated with the search color pattern with reference to the first database and searching for content associated with the searched text with reference to the second database. And a content search apparatus.   It further comprises history management means for managing the score based on the usage frequency of the searched content as a usage history database associated with the text having meanings related to the plurality of types of color patterns and the plurality of types of content. The content search apparatus according to claim 1.   3. The content search apparatus according to claim 2, wherein the search means is means for referring to the usage history database and reflecting the score in a search result of the content.   3. The content search apparatus according to claim 2, wherein the history management means is means for managing the usage history database in units of users.   The history management means estimates a score of the unknown text by referring to a score of another user other than the specific user when there is a text whose score is unknown in the usage history database for the specific user. The content search apparatus according to claim 4, wherein the content search apparatus is a means for performing the processing.   The search means is means for specifying a user who has input the search color pattern and reflecting the use history in a search result of the content with reference to the use history database of the specified user. The content search device according to claim 4 or 5. An analysis means for analyzing a synonym of the retrieved text;
The content search apparatus according to any one of claims 1 to 6, wherein the search means is means for searching for the content corresponding to the searched text and a text represented by a synonym thereof.   The first database associates color patterns with text in a first language, and the second database associates text in a second language different from the first language with content. 8. The apparatus according to claim 1, further comprising a translation unit that translates the retrieved text of the first language into the text of the second language. The content search device described. Accepts input of search color pattern,
With reference to a first database in which a plurality of types of color patterns and texts having meanings related to the respective color patterns are associated with each other, a search is performed for texts associated with the search color patterns. A content characterized in that a content associated with the retrieved text is searched with reference to a second database that associates the content with at least one text having a meaning associated with each content. retrieval method. A procedure for accepting input of a search color pattern;
With reference to a first database in which a plurality of types of color patterns and texts having meanings related to the respective color patterns are associated with each other, a search is performed for texts associated with the search color patterns. A step of searching for a content associated with the retrieved text with reference to a second database that associates the content with at least one text having a meaning associated with each content. A program for causing a computer to execute a characteristic content search method.

Images