JP5196569B2 - Content search device, content search method and program - Google Patents

Description

  The present invention relates to a content search apparatus, a content search method, and a program for searching a content by designating a tag from a content database to which a keyword representing the content content is assigned as a tag.

  In particular, content presenting a tag selection device that allows a user to easily select N tags of interest in a tag cloud in which tags that are conceptually related to each other are arranged close to each other in a two-dimensional space The present invention relates to a search device, a content search method, and a program.

  With the increase in the amount of information on the Internet and the advancement of search engine technology, various information can be acquired from the Internet. However, in order to select the information required by the user from the vast amount of information, it is necessary for the user to appropriately enter keywords for search, and the burden on the user to think about the keywords is small. Absent. In the first place, it is inevitable to input characters from the keyboard when inputting keywords.

  In recent years, in order to reduce the burden on these users as much as possible, by setting metadata consisting of short words and phrases representing the content of the searched content as tags, Attempts have been made to make it easy to get to the information you want.

  For example, in the photo sharing site Flickr (registered trademark) and the YouTube (registered trademark) of the moving image sharing site, keywords, categories, and genres expressing the contents of the photos and moving images can be assigned as tags. Moreover, the blog author can attach a tag corresponding to the content for each blog article. Further, in the case of a social bookmark Hatena Bookmark (registered trademark), a user can give a tag to the bookmark.

  There is a tag cloud as an interface for accessing tagged content. This “tag cloud” displays a set of tags having hyperlinks to tag search using tags as search conditions in a rectangular area. In this case, by changing the font size of the tag according to the popularity of the content, it is possible for the user to grasp which content is popular at a glance.

  In addition, when trying to navigate a large number of contents in a conventional tag cloud, the number of tags to be displayed is large (that is, the area of the tag cloud becomes large). For example, a general resolution display has about 500 tags. It is the limit to display the tag cloud that contains.

  However, in order to classify various contents, it is considered that 5000 to 10000 types of tags are necessary. If 5000 to 10000 types of tags are displayed as a tag cloud, there is a problem that it is difficult to find an appropriate tag from the displayed tag cloud. That is, many of the current tag clouds are arranged in alphabetical order or character code order, and it is difficult to find an interesting tag from among them.

  To solve this problem, an interface called a tag map has been proposed (for example, see Non-Patent Document 1). The “tag map” displays a plurality of tags conceptually related to each other in a two-dimensional space, and expresses the structural relationship between the tags by the shading or coloring of the background image of the tag map. As a result, the user can easily select a tag of interest.

The tag cloud or tag map allows the user to simply click on a tag displayed on the tag cloud / tag map via a mouse or Wii (registered trademark) remote control of Nintendo (registered trademark) Co., Ltd. The content to which the tag is attached can be searched, and the work of the user inputting the keyword for the search can be reduced.
Kan Fujimura “Blog mining and visualization technology equipped with BLOGRANGER TG”, “Artificial intelligence and knowledge processing” workshop / IEICE, AI2008-10, pp.57-62, 2008/6/30

  In the above prior art, only one tag can be used as a search condition. Assuming navigation of enormous amounts of information, there is a problem that sufficient selection cannot be made because the number of contents associated with one tag is large.

  The present invention provides a content search apparatus and method that allows a user to easily select N tags as search conditions when selecting content of interest using a tag map composed of a large number of tags. The purpose is to provide.

The content search apparatus of the present invention is designated by a tag map display means for displaying tag maps in which tags that are conceptually related to each other are arranged close to each other in a two-dimensional space, and a pointing device on the tag map. Pointing position acquisition means for acquiring the position, and N area tags (N is 1 or more) that are tags existing in a predetermined area based on the acquired pointing position, Depending on the means and the acquired N tags, if the closely related tags are selected using the similarity data existing behind the tags, an AND search search query is generated, and the closely related tags Is selected, a search query generating means for generating a search query for OR search and a keyword representing the contents are assigned as tags. The content database, sends the search query the search query generation means is generated, characterized by having a content retrieval means for retrieving the content.

  According to the present invention, when selecting a content of interest using a tag map composed of a large number of tags, there is an effect that a user can easily select N tags as a search condition. .

  The best mode for carrying out the invention is the following examples.

  FIG. 1 is a block diagram showing the overall configuration of a content search apparatus 100 that is Embodiment 1 of the present invention.

  The content search device 100 is a device that searches for content from a content database configured by a content set to which a tag that is a classification keyword is assigned using a tag as a search condition.

  The content search apparatus 100 includes tag statistical information 10, tag map generation means 20, tag map 30, user interface means 40, in-region tag acquisition means 50, and content search means 60.

  The tag map generation means 20 creates a tag map based on the tag statistical information 10 and stores it in the tag map 30. The tag map displays N tags that are conceptually related to each other in a two-dimensional space, and expresses the structural relationship between the tags by shading or coloring the background image of the tag map. .

  The user interface unit 40 includes a tag map display unit 41, a pointing position acquisition unit 42, a marker display unit 43, and a search result display unit 44.

  The tag map display means 41 displays the tag map 30 on the user terminal.

  The pointing position acquisition means 42 acquires a pointing position that is a position designated by the user with a pointing device such as a mouse on the tag map 30 and sends it to the server side.

  The in-area tag acquisition unit 50 uses the coordinates on the tag map 30 acquired by the pointing position acquisition unit 42 as a reference point, acquires N tags existing in the vicinity of the reference point, and stores them in the storage device. To do.

  The marker display unit 43 displays (provides) the markers for the N tags acquired by the in-region tag acquisition unit 50 on the tag map 30.

  The content search unit 60 includes a query generation unit 61, a content search unit 62, and a content database 63.

  The query generation means 61 generates a search query using the N tags selected by the in-region tag acquisition means 50 as a search condition, and stores it in the storage device.

  The content search unit 62 searches for the content in the content database 63 using the search query generated by the query generation unit 61 and stores it in the storage device.

  The search result display unit 44 displays the search result acquired by the content search unit 62 on the user screen.

  Note that the content in the above embodiment includes physical space objects to which tags can be attached, such as Web pages on the Internet, digital content such as images, sounds, and moving images, products, and services.

  FIG. 2 is a diagram illustrating the tag map 31 and the search image on the photo sharing site in the first embodiment.

  In other words, in FIG. 2, a photograph posted on a photo sharing site is searched, and two tags near the pointing position are selected, and the photo search result for the first tag and the photo for the second tag are displayed. The search results and the photo search results (with AND conditions) related to the two tags are displayed.

  Next, detailed operations in each means will be described.

  The tag map generation means 20 generates a tag map 30 by the method shown in Non-Patent Document 1. In Non-Patent Document 1, the Z coordinate of the tag is calculated, but the Z coordinate may not be used.

  The tag display means 41 displays the tag map 30 on the user terminal. However, since the tag map is generally huge, the tag cannot be read if the entire area is displayed. Therefore, the tag map is divided and the divided tag map (only a part of the tag map) is displayed. At this time, the tag map is scrolled according to an operation by the user's pointing device.

  Specifically, HTML representing the display area is created and sent to the Web browser of the terminal device. The Web browser acquires the image described in the HTML document from the server and displays it.

  FIG. 3 is a diagram illustrating a display example of the tag map 32 used in the first embodiment.

  The pointing position acquisition means 42 recognizes the user's pointing operation by embedding Java (registered trademark) Script in the HTML document sent to the user terminal, and acquires its coordinates. The acquired coordinates are sent to the server side by a function called Ajax of the asynchronous communication means.

  The in-area tag acquisition means 50 acquires N in-area tags, which are tags existing in a predetermined area centered on the pointing position, by the following procedures (1) to (4), and stores them in the storage device. Remember. N is an arbitrary number of 1 or more.

  (1) Draw a rectangular area with the coordinates of the pointing position transmitted from the pointing position acquisition means 42 as a reference point. In this case, for example, two points are provided at positions separated by a predetermined distance in two directions (directions different from each other by 180 degrees) from the reference point, and a rectangular region is formed so that these two points are diagonal points. Note that other methods may be employed as a method of forming the rectangular region based on the reference point.

  (2) The tag map 30 is accessed, and an in-area tag whose X coordinate and Y coordinate are within the rectangular area is extracted and stored in the storage device.

  (3) The distance between the pointing position and each extracted tag is calculated as the Euclidean distance.

  (4) N tags are selected in order of increasing Euclidean distance.

  FIG. 4 is a diagram illustrating an example of information registered in the tag map 30 in the first embodiment.

  As shown in FIG. 4, the tag map 30 holds tag IDs, tag names, X coordinates, Y coordinates, Z coordinates, and the like as data associated with tag units. Here, the tag ID is an identifier given uniquely for each tag.

  In FIG. 5, in the first embodiment, the pointing position designated by the pointing device is detected on the tag map 32 displayed on the user terminal, and the rectangular area R1 is drawn around the detected pointing position. FIG.

  FIG. 5 is a diagram illustrating an example in the case where one tag (gourmet) exists in the rectangular region R1.

  FIG. 6 is an image diagram in which a new location on the tag map is pointed and a rectangular region R2 is drawn around the pointing position in the first embodiment.

  The marker display means 43 attaches the marker M1 on the tag map 32 displayed on the user terminal regarding the tag acquired by the in-area tag acquisition means 50. Thus, the marker M1 is displayed on the tag map 32 displayed on the user terminal. A graphic other than the dumpling graphic may be displayed as the marker M1.

  FIG. 7 is a diagram illustrating an example in which a marker M1 corresponding to a tag (gourmet) included in the rectangular region R1 is displayed.

  FIG. 8 is a diagram illustrating a display example of the markers M2, M3, and M4 corresponding to the tags included in the rectangular region R2.

  The query generation unit 61 generates a query using the tag acquired by the in-region tag acquisition unit 50 as an AND condition. In the case illustrated in FIG. 8, a query for AND search of an “Italian” tag corresponding to the marker M2, a “bar” tag corresponding to the marker M3, and a “restaurant” tag corresponding to the marker M4 is generated.

  In this case, a query for performing an OR search for the acquired tag may be generated instead of a query for performing an AND search for the acquired tag.

  Or you may make it generate | occur | produce a search query according to the selected tag using the similarity data which exists behind a tag. For example, an AND search query may be generated if tags that are closely related are selected, and an OR search query may be generated if tags that are not closely related are selected.

  Note that “tags that are closely related using similarity data” are different from “tags that are close on the tag map”.

  Next, an example of generating a search query using similarity data will be described.

  FIG. 9 is a diagram illustrating a registration example of the tag similarity database.

  In FIG. 9, “is-a relationship” refers to a relationship in which tag ID: 02 is included in tag ID: 01 in the row of ID 1, and “member-of relationship” It is the relationship shown. Further, a relationship other than these may exist. For example, other relationships such as “has-a” and “part-of” may exist. In the case of a company tag map, relationships such as “partnership”, “parent company”, and “subsidiary” can be considered.

  FIG. 10 is a diagram illustrating an example of a tag map displaying similarity data (only the relationship) existing behind the tag, and is a diagram illustrating an example when the tag “player Y” is pointed.

  In the region R3 based on the pointing position, tags such as “J League”, “Team α”, “Team β”, “Major League”, and “Player X” exist. At this time, tags close to the tag “player Y” at the Euclidean distance are the tags “major league”, “team α”, and “player X”. On the other hand, by referring to the tag similarity data shown in FIG. 9, the tag “player Y” has a similar network distance of 1 with the tag “major league” and a similar network distance of 2 with the tag “player X”. Yes, the similar network distance with other tags is more than 2.

  In FIG. 10, tags connected by links are similar to each other. Information about each link is stored in the database shown in FIG. For example, in FIG. 10, since the tag “sports” (ID: 01) and the tag “baseball” (ID: 02) are linked by a link, the tag “sports” (ID: 01) and the tag “baseball” ( ID: 02) are similar to each other. This information is described in a record (row) with ID = 1 in FIG.

  That is, when viewed from the tag “player Y”, the tags that are close in similarity are “major league” and the tag “player X”.

  For example, from the tag “player Y”, select the tag “major league” that has the closest similarity as a tag with a close relationship, and generate a query that performs an AND search on the tag “player Y” and the tag “major league”. A case is considered. In addition, the tag “player X” having a similar network distance of 2 when viewed from the tag “player Y” is selected as a relatively unrelated tag, and an OR search query between the tag “player Y” and the tag “player X” is performed. It may be generated.

  FIG. 11 is a diagram showing an example of a tag map displaying similarity data (only the relationship) existing behind the tag, and an example of pointing between the tag “player Y” and the tag “team α” FIG.

  A tag “player Y” (ID: 08) and a tag “team α” (ID: 13) exist in the region R4 based on the pointing position. By referring to the tag similarity data shown in FIG. 9, it can be seen that these tags are separated by a similar network distance (low similarity). For this purpose, an OR search query for “player Y” and “team α” may be generated.

  FIG. 12 is a diagram showing an example of a tag map displaying similarity data (only the relationship) existing behind the tag, and is a diagram showing an example of pointing between the tag “player Y” and the tag “major league”. is there.

  In the region R5 based on the pointing position, a tag “player Y” (ID: 08) and a tag “major league” (ID: 07) exist. By referring to the tag similarity data shown in FIG. 9, it is understood that these tags have a similar network distance (high similarity). Therefore, an AND search of the tag “player Y” and the tag “major league” is performed. A query may be generated.

  In the above example, the number of selected tags is one or two, but the number of selected tags may be three or more N. For example, when three tags (tag “A”, tag “B”, tag “C”) are selected, tag “A” AND / OR tag “B” AND / Search may be performed as OR tag “C”, or a more complicated search query may be generated as (tag “A” AND tag “B”) OR tag “C”.

  The content search unit 62 accesses the content database 63 and executes the query generated by the query generation unit 61 to search for content and store it in the storage device.

  The search result display unit 44 transmits the search result of the content search unit 62 to the user terminal as a response to the user's tag selection operation.

  FIG. 13 is a diagram illustrating an example of content registered in the content database 63 in the first embodiment.

  For example, del. icio. Like a social bookmark service such as us, there is bookmark information to which a tag is attached when a user bookmarks a Web page. In the blog article, since a keyword is specified in the <dc: subject> field of RSS, the blog article to which such RSS is given is an example of content to which a tag is attached.

  As shown in FIG. 13, the content database 63 stores content IDs, tags, contents, and the like as data associated with content units. Here, the content ID is an identifier uniquely given to each content. A tag is a word or phrase that briefly indicates the content of the text set by the document creator. The content is the content itself. Also, as shown in FIG. 13, additional information such as URL, registration date and time, title, etc. may be added to hold the data.

  In addition, in this way, instead of explicitly assigning a tag, the user's name, organization name, place name specific expression, etc. included in the text in the content are extracted and characteristic from these Those appearing in may be automatically extracted as tags. A set of texts to which a tag is explicitly assigned is used as learning data, and a tag is automatically assigned to text that is not assigned a tag according to the degree of similarity with the text that is assigned a tag. Also good.

  FIG. 14 is a block diagram showing a content search apparatus 200 that is Embodiment 2 of the present invention.

  The content search apparatus 200 is an apparatus provided with user interface means 40 a instead of the user interface means 40 in the content search apparatus 100.

  The user interface unit 40 a is obtained by adding a marker control unit 45 between the marker display unit 43 and the search result display unit 44 in the user interface unit 40.

  The marker control unit 45 executes processing such as moving, copying, or deleting the marker attached to the tag. When moving a marker, select the marker attached to the tag with the pointing device (if you are using a mouse, grab it with the mouse), and move the marker to another tag you want to attach the marker. Then, by releasing the marker at that location, the marker is moved, that is, dragged.

  When deleting a marker, if the pointing device is a mouse, the marker is deleted by right-clicking on the marker attached to the tag and selecting <Delete> from the displayed menu.

  FIG. 15 is a diagram illustrating an example of a screen after the user has moved the marker M3.

  Content search processing by the content search means 60 is performed at the timing when the marker is moved to a new tag.

  According to the second embodiment, when a search condition is changed only by an operation with a pointing device, the search condition can be changed more visually, so that the user is relieved from the stress of determining and inputting a keyword. .

  FIG. 16 is a block diagram showing a content search apparatus 300 that is Embodiment 3 of the present invention.

  The content search apparatus 300 is an apparatus provided with user interface means 40b instead of the user interface means 40 in the content search apparatus 100.

  The user interface unit 40 b is obtained by adding a similarity data display unit 46 between the marker display unit 43 and the search result display unit 44 in the user interface unit 40.

  The similarity data means 46 displays the similarity data existing behind the tag to which the marker is attached.

  FIG. 17 is a diagram illustrating an example of a tag map displaying similarity data.

  In FIG. 17, a marker is attached to the tag “brand A”. This is an example in which a tag similar to the tag “brand A” is found by accessing the tag similarity database shown in FIG. 9 and a link and relevance are displayed between the tags.

  FIG. 18 is a diagram illustrating an example of a tag map displaying similarity data.

  In FIG. 18, similarity data related to the tag “mochi shochu” is displayed. At this time, the tag information database shown in FIG. 4 is accessed to obtain the abstract level of similar tags. The display method of the similarity data may be changed according to the level of abstraction with respect to the reference tag (“Shochu shochu” in FIG. 18). In FIG. 18, similarity data for a tag having a higher abstraction level than the reference tag is indicated by a solid line, and similarity data for a tag having a lower abstraction level is indicated by a dotted line.

  The “abstract level of similar tags” is the Z coordinate of the tag described in Non-Patent Document 1, and if the Z coordinate is not used, the “abstract level of similar tags” is not acquired. .

  FIG. 19 is a diagram illustrating an example in which only similarity data with a tag having a higher abstraction level than the reference tag is displayed.

  FIG. 20 is a diagram illustrating an example in which only similarity data with a tag having a lower abstraction level than the reference tag is displayed.

  As shown in FIGS. 19 and 20, the display may be switched according to the degree of abstraction with respect to the reference tag.

  FIG. 21 is a diagram illustrating an example in which only the similarity between tags is displayed (relationship is not displayed).

  As shown in FIG. 21, it may be possible to change the similarity data to be displayed.

  Further, all similar data may be displayed without providing the reference tag. Further, the contents to be displayed may be determined by combining the above.

  In addition, the marker may be moved in the tag map displaying the similarity data. That is, a marker moving means for changing the tag to be selected by selecting a marker displayed on the tag map with a pointing device and moving it to another tag may be provided.

  In the above embodiment, a mouse is used as an example of the pointing device, but a pointing device such as a touch panel may be used in addition to this. The mouse click is an example of pointing, and the mouse click position is an example of a pointing position.

  In the above embodiment, when a predetermined area is set, a shape other than a rectangle, such as a polygon other than a rectangle or a circle, may be set instead of a rectangle.

  According to the above embodiment, a pointing position (mouse click position) on the tag map is acquired, a marker is attached to an in-area tag existing in a predetermined area centered on the pointing position, and N tags are added. Since the user can easily select the contents, the user can perform an AND / OR search on the contents related to the N tags by an intuitive operation using a pointing device such as a mouse.

  That is, in the above embodiment, tag map generating means for generating a tag map in which tags conceptually related to each other are arranged close to each other in a two-dimensional space, and tag map display means for displaying the tag map; , Pointing position acquisition means for acquiring a position designated by the pointing device on the tag map, and N in-area tags that are tags existing in a predetermined area based on the acquired pointing position In-area tag acquisition means, marker display means for displaying a marker indicating that the search is to be performed on the N in-area tags acquired by the in-area tag acquisition means, and search using the selected tag as a condition It is an example of the content search apparatus which has a search query production | generation means which produces | generates a query.

  Note that there may be one in-region tag or a plurality of in-region tags that are present in a predetermined region based on the acquired pointing position. Even if there is only one in-area tag, the above embodiment is characterized in that a tag (in-area tag) existing in a predetermined area corresponding to the pointed position is obtained. This point is different from the conventional example. It is a different point.

  In this case, the search query generating means is means for generating a search query using similarity data existing behind the selected tag. Further, it may have a content search means for searching the content by sending the search query generated by the search query generation means to a content database to which a keyword representing the content content is assigned as a tag. Furthermore, you may make it have a search result display means to display a search result. In addition, similarity data display means for displaying similarity data that exists behind the tag with marker displayed on the tag map may be provided.

  Moreover, the said Example can be grasped | ascertained as invention of a method. In other words, if the means in the above embodiment is replaced with a process, the above embodiment is an example of a method invention. That is, the above embodiment generates a tag map in which tags conceptually related to each other are arranged close to each other in a two-dimensional space, and stores the tag map in the storage device, and displays the tag map. A tag map display step, a pointing position acquisition step for acquiring a position designated by the pointing device on the tag map, and a tag existing in a predetermined region based on the acquired pointing position An in-area tag acquisition step for acquiring N tags and storing them in a storage device, and a marker indicating that the N-in-area tag acquired in the in-area tag acquisition step is a search target And a marker display step to be displayed on the screen and a search query generation step of generating a search query using the selected tag as a condition and storing the search query in a storage device. It is an example of a search method.

  Moreover, the above embodiment can be grasped as a program invention. That is, the said Example is an example of the program which makes a computer function as each means which comprises the said content search apparatus.

  That is, the said Example is an example of the program which makes a computer implement | achieve the said content search method.

  Moreover, the above embodiment can be grasped as an invention of a recording medium. That is, the above-described embodiment is an example of a computer-readable recording medium that records the above program. As a recording medium, a CD, a DVD, an FD, an HD, an optical disk, a magneto-optical disk, a semiconductor memory, and the like can be considered.

1 is a block diagram illustrating an overall configuration of a content search apparatus 100 that is Embodiment 1 of the present invention. In Example 1, it is a figure which shows the tag map 31 and the search image in a photograph sharing site. It is a figure which shows the example of a display of the tag map 32 used in Example 1. FIG. In Example 1, it is a figure which shows the example of the information registered into the tag map. In the first embodiment, the pointing position designated by the pointing device is detected on the tag map 32 displayed on the user terminal, and the rectangular region R1 is drawn with the detected pointing position as the center. It is an image figure shown. In Example 1, it is the image figure by which the new location on a tag map is pointed and the rectangular area | region R2 is drawn centering on this pointing position. It is a figure which shows the example which is displaying the marker M1 corresponding to the tag (gourmet) contained in rectangular area R1. It is a figure which shows the example of a display of the marker M2, M3, M4 corresponding to the tag contained in rectangular area R2. It is a figure which shows the example of registration of a tag similarity database. It is a figure which shows the example of the tag map which displayed the similarity data (only the relationship) which exists behind a tag, and is a figure which shows the example at the time of pointing the tag "player Y". It is a figure which shows the example of the tag map which displayed the similarity data (only the relationship) which exists behind a tag, and is a figure which shows the example at the time of pointing between the tag "player Y" and the tag "team α" is there. It is a figure which shows the example of the tag map which displayed the similarity data (only the relationship) which exists behind a tag, and is a figure which shows the example which pointed between the tag "player Y" and the tag "major league". 6 is a diagram illustrating an example of content registered in a content database 63 in Embodiment 1. FIG. It is a block diagram which shows the content search apparatus 200 which is Example 2 of this invention. It is a figure which shows the example of the screen after the user moves the marker M3. It is a block diagram which shows the content search apparatus 300 which is Example 3 of this invention. It is a figure which shows the example of the tag map which displayed the similarity data. It is a figure which shows the example of the tag map which displayed the similarity data. It is a figure which shows the example which displays only the similarity data with a tag with a high abstraction level with respect to a reference | standard tag. It is a figure which shows the example which displays only the similarity data with a tag with a low abstraction level with respect to a reference | standard tag. It is a figure which shows the example which displays only the similarity between tags (relation is not displayed).

Explanation of symbols

100: Content search device,
10: Tag statistics information,
20 ... tag map generation means,
30 ... Tag map,
40. User interface means,
41 ... tag map display means,
42: Pointing position acquisition means,
43 ... Marker display means,
44 ... Search result display means,
50: In-area tag acquisition means,
60: Content search means,
61 ... Query generation means,
62 ... content search means,
63 ... Content database,
200 ... content search device,
45 ... marker control means,
40a ... User interface means,
300 ... content search device,
40b ... User interface stage,
46: Similarity data display means.

Claims (4)

Tag map display means for displaying tag maps in which tags conceptually related to each other are arranged close to each other in a two-dimensional space;
Pointing position acquisition means for acquiring a position designated by the pointing device on the tag map;
An intra-area tag acquisition means for acquiring N intra-area tags (N is 1 or more), which are tags existing in a predetermined area based on the acquired pointing position;
According to the acquired N tags, if similar tags are selected using the similarity data existing behind the tags, a search query for AND search is generated, and tags that are not closely related are selected. Search query generation means for generating a search query for OR search, if any ;
Content search means for searching the content by sending the search query generated by the search query generation means to a content database to which a keyword representing the content content is assigned as a tag;
A content search apparatus comprising: In claim 1,
Marker display means for displaying a marker indicating that it is a search target on the N intra-area tags acquired by the intra-area tag acquisition means;
Similarity data display means for displaying similarity data existing behind the tag with marker displayed on the tag map;
Content search device characterized in that it comprises a. A tag map display step for displaying tag maps in which tags conceptually related to each other are arranged close to each other in a two-dimensional space;
A pointing position acquisition step of acquiring a position designated by the pointing device on the tag map;
An in- area tag acquisition step of acquiring N in-area tags (N is 1 or more) that are tags existing in a predetermined area with the acquired pointing position as a reference;
According to the acquired N tags, if similar tags are selected using the similarity data existing behind the tags, a search query for AND search is generated, and tags that are not closely related are selected. A search query generation step for generating a search query for OR search, if any;
A content search step of sending the search query generated in the search query generation step to a content database to which a keyword representing the content content is assigned as a tag and searching for the content;
A content search method characterized by comprising: A program for causing a computer to function as each means constituting the content search device according to claim 1 .