Data search device, data search method and data search program

Abstract

To provide a data retrieval device, a data retrieval method and a data retrieval program that can easily find a target retrieval object in retrieval.SOLUTION: A tag item storage unit 11 stores correspondence between tags and data to which the tags are added. A tag matrix storage unit 13 stores a tag matrix which has elements indicated with row tags and column tags and displays information related to data to which corresponding row tags and column tags are added, element by element. A tag matrix output unit 15 causes an output device 3 to display the tag matrix. When a specific element is selected from elements included in the tag matrix, a low-order classification output unit 16 performs first classification of first data to which a first row tag and a first column tag corresponding to the specific element are added by using a plurality of second tags other than the first row tag and first column tag to generate a plurality of first groups, changes states of display areas according to feature quantities of the respective first groups, and causes the display device to display the respective first groups.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a data search device, a data search method, and a data search program. [Background technology]

[0002] There are known data sharing systems that store and share various types of electronic information. In particular, there are known data sharing systems in which registered users can freely register data via a network and freely view the data registered in a database. Because such data sharing systems store a variety of information, it is important to manage the classification of information.

[0003] For example, when storing and managing documents, tagging them with arbitrary character strings makes it possible to retrieve the desired document using the tag. In this case, documents also include displayable data. However, documents contain various elements, so it is not realistic to classify documents with a single tag. For this reason, documents are often assigned multiple tags to classify and manage them.

[0004] Furthermore, when searching for a desired document, it can be difficult to find it if one searches based on vague memories. Therefore, there are technologies that make it easier to find a desired document even with vague memories by searching by digging into key keywords such as tag information. For example, there is a search technology that uses a matrix table that cross-tabulates two pieces of tag information. There is also a technology that analyzes documents by combining qualitative data such as text, images, and audio with quantitative data such as numerical and categorical data. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2014-177758 [Patent Document 2] Japanese Patent Application Laid-Open No. 2006-269802 [Patent Document 3] Japanese Patent Application Laid-Open No. 2000-83491 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the number of documents is increasing every day, and it is difficult to find the desired item among the increasing number of documents by two-dimensional cross-tabulation or simple keyword digging. Furthermore, while technology that combines qualitative and quantitative data for analysis can perform appropriate classification, it is difficult to improve search efficiency.

[0007] In one aspect, the present invention aims to provide a data search device, a data search method, and a data search program that enable a user to easily find a desired search target during a search. [Means for solving the problem]

[0008] In a first proposal, a data search device searches for data stored in a database using row tags and column tags, which are arbitrary character strings. A tag item storage unit stores a correspondence between the row tags and column tags and data to which the row tags and column tags are attached. A tag matrix storage unit stores a tag matrix having elements indicated by the row tags and column tags, and displaying information about the data to which the row tags and column tags corresponding to each element are attached. A tag matrix output unit displays the tag matrix on a display device. When a specific element to which a first row tag and a first column tag are attached and which corresponds to first data is selected from the elements included in the tag matrix displayed on the display device, a sub-classification output unit identifies a predetermined number of second tags, among tags other than the first row tag and the first column tag, that are highest in total number attached to the first data, and performs a first classification of the first data using the predetermined number of second tags according to the characteristics of each of the first data, thereby generating a plurality of first groups.When a specific first group is designated by a predetermined operation from among the generated first groups, a list of items, which are the first data and to which the second tag is added and which are included in the designated specific first group, is arranged near the multiple first groups. The display device displays the image. [Effects of the Invention]

[0009] When searching, you can easily find what you are looking for. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram of a data search device according to a first embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the data structure of tag data. [Figure 3] FIG. 3 is a diagram illustrating an example of the data structure of a tag item. [Figure 4] FIG. 4 is a diagram illustrating an example of the data configuration of the tag matrix. [Figure 5] FIG. 5 is a diagram showing an example of a tag matrix display screen that displays a tag matrix as thumbnails. [Figure 6] FIG. 6 is a diagram showing an example of a tag matrix display screen that displays a selected tag matrix. [Figure 7] FIG. 7 is a diagram showing an example of a tag matrix display screen that displays related information for a selected tag combination. [Figure 8] FIG. 8 is a diagram illustrating a display example of a tag matrix display screen that displays classifications from the third tag onward according to the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of the item display screen. [Figure 10] FIG. 10 is a diagram showing an example of another method for displaying items. [Figure 11] FIG. 11 is a flowchart of the new tag addition process. [Figure 12] FIG. 12 is a flowchart of a search process according to the first embodiment. [Figure 13] FIG. 13 is a diagram illustrating a display example of a tag matrix display screen that displays classifications from the third tag onward according to the second embodiment. [Figure 14]FIG. 14 is a flowchart of a search process according to the second embodiment. [Figure 15] FIG. 15 is a diagram illustrating a display example of a tag matrix display screen that displays classifications from the third tag onward according to the third embodiment. [Figure 16] FIG. 16 is a diagram illustrating a display example of a tag matrix display screen that displays classifications from the third tag onward according to the fourth embodiment. [Figure 17] FIG. 17 is a diagram showing an example of the hardware configuration of a computer that realizes the same functions as the data search device.

[0011] The following describes in detail embodiments of the data search device, data search method, and data search program disclosed in the present application with reference to the accompanying drawings. Note that the data search device, data search method, and data search program disclosed in the present application are not limited to the following embodiments. Example 1

[0012] 1 is a block diagram of a data search device according to a first embodiment. The data search device 1 is realized by a computer including hardware such as a CPU (Central Processing Unit) and memory, and a software program. The data search device 1 includes a tag item storage unit 11, a tag matrix editing unit 12, a tag matrix storage unit 13, an item number counting unit 14, a tag matrix output unit 15, an item output unit 17, a display log storage unit 18, a tag editing unit 19, and a sub-classification output unit 16.

[0013] Furthermore, the data search device 1 is connected to an input device 2 and an output device 3. The data search device 1 is also connected to a database 4. The input device 2 is, for example, a keyboard or a mouse. The output device 3 is, for example, a display or a printer.

[0014] Database 4 is a data storage system that stores various data called items registered by users who are authorized to use the database. Examples of the various data include documents, images, binary data such as video or audio, blogs, URLs, etc.

[0015] The tag item storage unit 11 stores tag data and tag items. Here, the tag data and tag items will be described.

[0016] Tag data is data that indicates information about tags added to items stored in the database 4. Tag data is newly generated when a new word or phrase is registered as a tag. Tag data includes items such as tag ID, tag word, number of registrations, number of views, creation date and time, update date and time, and deletion date and time. Fig. 2 is a diagram showing an example of the data structure of tag data.

[0017] "Tag ID" is information that uniquely identifies a tag stored in the tag item storage unit 11. "Tag word" is an arbitrary string of characters such as a word or phrase that is a tag added to an item in the database 4. "Number of registrations" is the total number of tag items to which the same tag ID is set. "Number of views" is the total number of times a tag corresponding to a tag ID has been output to the output device 3. "Creation date and time" is the date and time when the tag data was created. "Update date and time" is the date and time when the contents of the tag data were updated. "Deletion date and time" is the date and time when the tag data deletion operation was performed.

[0018] Furthermore, as shown in FIG. 2, the tag data may have the following items: number of registrations (within the last 3 months), number of registrations (within the last 6 months), number of registrations (within the last 1 year), number of views (within the last 3 months), number of views (within the last 6 months), and number of views (within the last 1 year). The "number of registrations (within the last 3 months)" indicates the total number of tag items registered within the past 3 months from the present. Furthermore, the "number of registrations (within the last 6 months)" and the "number of registrations (within the last 1 year)" indicate the total number within the past 6 months and the past 1 year from the present, respectively. The same applies to the "number of views (within the last 3 months)," the "number of views (within the last 6 months)," and the "number of views (within the last 1 year)." The number of registrations and the number of views for each period are used to obtain information showing changes over time.

[0019] Tag data is generated by the tag editing unit 19 when a new word or phrase is input as a tag. The tag data is also updated every time a word or phrase identical to the tag word is added to or deleted from an item, or every time an item with that tag is viewed.

[0020] A tag item is data indicating the correspondence between an item registered in the database 4 and a tag attached to that item. A tag item is uniquely associated with tag data by a tag ID. A tag item is generated by the tag editing unit 19 when a new tag is registered, and is updated every time a corresponding tag is attached to a new item.

[0021] 3 is a diagram showing an example of the data structure of a tag item. A tag item stored in the tag item storage unit 11 has items such as a tag ID, an item ID, an item type, a creation date and time, an update date and time, and a deletion date and time.

[0022] "Tag ID" is identification information associated with tag data, and the tag ID of the tag data is used. "Item ID" is information that uniquely identifies the item to which the tag is attached, and "Item Type" is information that indicates the data type of the item. For each item to which a tag is attached, a pair of item ID and item type is registered. "Creation Date and Time" is the date and time when the tag item was created. "Update Date and Time" is the date and time when the contents of the tag item were updated. "Deletion Date and Time" is the date and time when the tag item was deleted.

[0023] An item such as a document is registered in the database 4 together with an item ID. The item ID associates the item with a tag item. The tag ID also associates the tag item with tag data. In this embodiment, the item and tag are associated via the tag item, but it is also possible to integrate the tag and tag item and directly associate the tag with the item.

[0024] The tag matrix editing unit 12 acquires two or more tag groups each containing one or more tags that are arbitrary character strings input by the user. The tag matrix editing unit 12 then generates a tag matrix in which tags contained in one of the acquired tag groups are row elements and tags contained in the other are column elements, and stores the generated tag matrix in the tag matrix storage unit 13.

[0025] 4 is a diagram showing an example of the data configuration of a tag matrix. The tag matrix stored in the tag matrix storage unit 13 has items such as a matrix ID, a column tag, a row tag, the number of items, creation date and time, update date and time, and deletion date and time.

[0026] "Matrix ID" is information that uniquely identifies a tag matrix stored in the tag matrix storage unit 13. "Column tag" is information that indicates tags included in one tag group entered by the user and is identified by a tag ID. When the tag matrix is displayed, each tag in the column tags becomes a tag item that makes up the X-axis. "Row tag" is information that indicates tags included in another tag group entered by the user and is identified by a tag ID. When the tag matrix is displayed, each tag in the row tags becomes a tag item that makes up the Y-axis. "Number of items" indicates the total number of items searched by search criteria, which are a combination of column tags and row tags that make up the tag matrix. "Creation date and time" is the date and time when the tag matrix was created. "Update date and time" is the date and time when the contents of the tag matrix were updated. "Deletion date and time" is the date and time when a deletion operation was performed on the tag matrix.

[0027] A tag matrix is a matrix whose elements are cells that are the intersections of column tags and row tags. The tag matrix is displayed with information about items to which column tags and row tags are added registered in each cell.

[0028] When the tag matrix editing unit 12 receives a request to generate a tag matrix from a user, it displays a tag matrix creation screen on the output device 3, waits for user input, and creates a tag matrix based on the input information. The tag matrix creation screen is a screen for inputting information related to the tag matrix. The tag matrix creation screen has areas for inputting, for example, the title, creation date, expiration date, action upon expiration, comment permission, and use of the tag matrix.

[0029] For each tag matrix stored in tag matrix storage unit 13, item number tallying unit 14 searches for tag items stored in tag item storage unit 11 using a combination of row tags and column tags corresponding to each cell (intersection) of the tag matrix as a search condition. Then, item number tallying unit 14 calculates the number of items assigned tags that match the tag combination and writes the calculated number of items into cell tally data corresponding to each cell. Item number tallying unit 14 stores the cell tally data in tag matrix storage unit 13 in association with the tag matrix used to calculate the number of items.

[0030] The cell summary data is generated to correspond to the number of cells in the tag matrix. The cell summary data may also include information on two corresponding tags and the number of item IDs found in the pair. This cell summary data is an example of "information on data tagged with row and column tags."

[0031] The tag matrix output unit 15 receives a request to display a tag matrix from a user via the input device 2. Then, the tag matrix output unit 15 retrieves the tag matrix requested to be displayed from the tag matrix storage unit 13. Furthermore, the tag matrix output unit 15 obtains cell totalization data associated with the retrieved tag matrix. Then, the tag matrix output unit 15 registers the numbers of the cell totalization data in each cell of the retrieved matrix and outputs the data to the output device 3.

[0032] Fig. 5 is a diagram showing an example of a tag matrix display screen that displays a tag matrix as thumbnails, Fig. 6 is a diagram showing an example of a tag matrix display screen that displays a selected tag matrix, and Fig. 7 is a diagram showing an example of a tag matrix display screen that displays related information for a selected tag combination.

[0033] The tag matrix display screen 101 shown in FIG. 5 is a screen that displays the tag matrix stored in the tag matrix storage unit 13, and has areas for a "frequently viewed tag matrix" and a "recently viewed tag matrix." The "frequently viewed tag matrix" is an area that displays the titles or thumbnails of tag matrices that have been viewed frequently. The "recently viewed tag matrix" is an area that displays the titles or thumbnails of tag matrices that have been viewed frequently in the most recent period. The number of views of a tag matrix can be determined from the viewing log, which will be described later.

[0034] The tag matrix display screen 102 shown in FIG. 6 shows a state in which a tag matrix selected by the user using the tag matrix display screen 101 shown in FIG. 5 is displayed.

[0035] In the tag matrix, for example, "portal site, virtualization, server, ..." is set in the "column tags," and "portal site, sales promotion, proposal, ..." is set in the "row tags." The tags (phrases) registered in the column tag and row tag items are then displayed in the tag matrix on tag matrix display screen 102. The cell at the intersection of each item in the tag matrix displays the number of items corresponding to the value of the cell aggregation data corresponding to the cell of the corresponding combination of column tag and row tag.

[0036] For example, the number of accesses "50" displayed in the cell at the intersection where the column tag is "portal site" and the row tag is "portal site" indicates that the number of items tagged with "portal site" among the items stored in database 4 is 50. Similarly, the number of accesses "10" displayed in the cell at the intersection where the column tag is "portal site" and the row tag is "sales expansion" indicates that the number of items tagged with "portal site" and "sales expansion" is 10.

[0037] Furthermore, the tag matrix output unit 15 receives an input of a selection of a cell (intersection) on the tag matrix displayed on the tag matrix display screen 102 from the input device 2. Then, the tag matrix output unit 15 acquires all tags attached to each item from all items extracted from the tag item storage unit 11 using a combination of a column tag and a row tag corresponding to the selected cell (intersection). Next, the tag matrix output unit 15 identifies, from the acquired tags, tags other than those set as column tags and row tags in the tag matrix as classification tags, and outputs a list of the identified classification tags. Furthermore, the tag matrix output unit 15 outputs a list of items extracted by a search process using a combination of a column tag and a row tag corresponding to the selected cell (intersection).

[0038] The tag matrix display screen 103 shown in Fig. 7 is a screen that is displayed when a predetermined first operation is performed on an intersection of the tag matrix on the tag matrix display screen 102 in Fig. 6. The first operation is, for example, a double-click. The tag matrix display screen 103 shows a state in which the first operation is performed on the cell at the intersection of "Portal Site" and "Sales Promotion", and a list of items is displayed with the number of items set to 10.

[0039] The tag matrix display screen 103 has display areas for the selected tag combination, a list of items, and a list of other tags. The "selected tag combination" is an area that displays the column tag and row tag corresponding to the cell (intersection) selected on the tag matrix displayed on the tag matrix display screen 102. The "item list" is an area that displays a list of items identified as the number of items written in the selected cell, i.e., items to which the corresponding column tag and row tag are attached. In the example shown in FIG. 7, the item title and creation date / update date are displayed as information identifying each item to which the two tags "portal site" and "sales promotion" stored in the database 4 are attached. Note that the "item list" may display a description of each item stored in the database 4 as related information about the item, along with the title and creation date / update date of each item. The "other tag list" is an area that displays a list of classification tags, which are tags other than the column tags and row tags of the tag matrix, among all tags attached to the cell (intersection) selected using the tag matrix display screen 102. Classification tags are tags other than the column tags and row tags attached to items.

[0040] When a tag is selected in the "Other Tag List" on the tag matrix display screen 103, the tag matrix output unit 15 narrows down the search for the items displayed in the "Item List" on the tag matrix display screen 103 by using the selected tag. Then, it extracts items to which the selected tag is attached and displays a list of those items on the tag matrix display screen 103. Furthermore, when another tag is selected in the "Other Tag List" on the tag matrix display screen 103, the tag matrix output unit 15 performs a similar narrowing down process using the selected tag. Then, when a specific item is selected from the items displayed in the "Item List" on the tag matrix display screen 103, the tag matrix output unit 15 notifies the item output unit 17 of information about the selected item.

[0041] Furthermore, each time the tag matrix is displayed, the tag matrix output unit 15 accumulates log information of the display process as a viewing log in the display log storage unit 18. The viewing log has items such as tag ID, user ID, and creation date and time. "Tag ID" is the tag ID of the tag attached to the viewed item. "User ID" is the user ID of the user who viewed the item. "Creation date and time" is information indicating the date and time when the viewing log was created. Furthermore, the tag matrix output unit 15 tallies the number of views for a predetermined period and stores and updates the number of views of each tag data in the display log storage unit 18.

[0042] The subclassification output unit 16 receives operation information input by a user using the input device 2 to the tag matrix display screen 102 shown in Fig. 6. When a specific cell is selected on the tag matrix display screen 102 by performing a predetermined second operation on the specific cell, the subclassification output unit 16 causes the output device 3 to display classifications from the third tag onwards. This selected cell is an example of a "specific element," and the row tag and column tag corresponding to the selected element are examples of a "first row tag" and a "first column tag."

[0043] 8 is a diagram showing a display example of a tag matrix display screen that displays classifications from the third tag onward according to the first embodiment. Here, the second operation is an operation different from the first operation, such as a left click. Furthermore, when the list display button 143 is pressed after the second operation for an object selected on the tag matrix display screen 104, the sub-classification output unit 16 causes the output device 3 to display items included in the selected object. Details of the display of the third tag onward and the item display will be described below.

[0044] The subclassification output unit 16 can obtain a combination of third tags corresponding to the selected cell in the tag matrix. There are no particular limitations on the method for obtaining the third tags, and various methods described below can be used.

[0045] For example, the subclassification output unit 16 may store in advance a combination of third tags for each combination of two tags corresponding to each cell in the tag matrix. Alternatively, the subclassification output unit 16 may select the top four tags in terms of the total number of tags assigned to each item included in the selected cell. The subclassification output unit 16 then classifies each item included in the selected cell into one of the tag groups of the third tags based on the assigned tag. For example, the subclassification output unit 16 may classify a specific item into the tag among the third tags that best represents the characteristics of the specific item. The item included in the selected cell based on this third tag is an example of "first data," the classification of the item included in the selected cell based on the third tag is an example of "first classification," and the tag group of the third tag is an example of "first group."

[0046] Next, the subcategory output unit 16 counts the number of items included in each third tag. Then, the subcategory output unit 16 determines the size of the display area for the third tag so that the display area becomes larger as the number of items increases. In this embodiment, the subcategory output unit 16 displays each tag group of the third tag in a stacked manner using a cylindrical graph. Then, the subcategory output unit 16 determines the display width of each tag group so that the width of the layer in the cylindrical display area, i.e., the length in the axial direction, becomes wider as the number of items increases. This number of items included in each third tag is an example of a "feature."

[0047] For example, a case will be described in which the second operation is performed on the cell at the intersection of "portal site" and "sales promotion" on the tag matrix display screen 104 in Fig. 8. In this case, the sub-classification output unit 16 displays a cylindrical graph 141 representing the classification of the third tag for the items displayed with the number of items=55 on the tag matrix display screen 104, as shown in Fig. 8. Then, the sub-classification output unit 16 displays, for example, the tag word and the number of items for each tag group.

[0048] In this case, the third tag #1 contains the largest number of items, so the tag group of the third tag #1 has the largest display width that can be placed in the cylindrical graph 141. The third tag #2 contains the next largest number of items, so the tag group of the third tag #2 has the next largest display width that can be placed in the cylindrical graph 141. The third tag #3 contains the next largest number of items, so the tag group of the third tag #3 has the next largest display width that can be placed in the cylindrical graph 141. The third tag #4 contains the fewest number of items, so the tag group of the third tag #4 has the smallest display width that can be placed in the cylindrical graph 141.

[0049] Next, when a second operation is performed on a tag group of a specific third tag from the cylindrical graph 141, the subclassification output unit 16 displays a cylindrical graph 142 representing the classification of the fourth tag on the tag matrix display screen 104 using the same process as for the third tag. In this case, the subclassification output unit 16 also increases the display width of the cylindrical graph 142 for each tag group in descending order of the number of items included in each tag group of the fourth tag. In this case, the tag group of the selected third tag is an example of a "selected first group." Furthermore, the classification using the fourth tag is an example of a "second classification," and the tag group of the fourth tag is an example of a "second group." The items included in each tag group of the fourth tag are an example of "second data."

[0050] The subclassification output unit 16 sequentially displays cylindrical graphs representing the classification of tags at the next layer in response to the input of the second operation until there are no more tags at the next layer.

[0051] Fig. 9 is a diagram showing an example of an item display screen. When the list display button 143 is pressed after the second operation or when there is no next hierarchical level, the sub-classification output unit 16 causes the output device 3 to display the item display screen 105 shown in Fig. 9. For example, when the third tag #2 in the cylindrical graph 141 in Fig. 8 is selected and the list display button 143 is pressed, the sub-classification output unit 16 causes the item display screen 105 to display information about each item included in the tag group of the third tag #2.

[0052] Here, there are no particular limitations on the information that can be displayed on the item display screen 105, as long as it is information related to the item. For example, the item display screen 105 may display the item title, the creation date / update date and time, and a description of the item that is stored in the database 4 as related information for the item.

[0053] Furthermore, in this embodiment, the items included in the selected target are displayed using the item display screen 105, which is a screen separate from the tag matrix display screen 104, but the display method is not limited to this. FIG. 10 is a diagram showing an example of another method of displaying items. For example, when a predetermined third operation is performed on a specific tag group, the subclassification output unit 16 may display a list of items 145 near the tag group that is the target of the third operation, as shown in FIG. 10. FIG. 10 shows the state when the third operation is performed on the tag group representing the fourth tag ##2 in the cylindrical graph 142. The third operation is an operation different from the second operation, such as a right-click.

[0054] When a specific item is selected from the items displayed on the item display screen 105, the sub-classification output unit 16 notifies the item output unit 17 of information about the selected specific item.

[0055] When an item is selected in the "item list" on the tag matrix display screen 103, the item output unit 17 retrieves the corresponding item from the database 4 using the item ID and displays it on the output device 3. When an item is selected on the item display screen 105, the item output unit 17 retrieves the corresponding item from the database 4 using the item ID and displays it on the output device 3.

[0056] The tag editing unit 19 adds and updates tags to be added to items in the database 4. Tags may be generated mechanically or may be input by the user.

[0057] In the case of mechanical generation, the tag editing unit 19 generates a learning model by learning to extract features using multiple items. Then, using the generated learning model, it is possible to extract features from items newly registered in the database 4 and use them as tags.

[0058] In addition, in the case of input by the user, when the user requests tag input, the tag editing unit 19 displays a tag input screen. The tag input screen includes areas such as the item title, registered tags, tag input, and areas for a register button and a clear button. Then, using the item information and tag information input by the user using the tag input screen, the tag editing unit 19 adds the input tag to the specified item.

[0059] If the input tag is an existing tag, the tag editing unit 19 adds an item ID and an item type to the tag item corresponding to the existing tag. Furthermore, the tag editing unit 19 retrieves tag data corresponding to the specified tag from the tag item storage unit 11, increments the number written in the registration count by 1, and rewrites the update date and time to the current date and time. On the other hand, if the input tag is a new tag, the tag editing unit 19 registers new tag data, and further generates a tag item corresponding to the tag and stores it in the tag item storage unit 11.

[0060] In addition, the tag editing unit 19 may detect variations in the spelling of tag words registered in the tag data at predetermined times or periods, check whether multiple spellings are mixed in tags that have the same meaning, and unify the spellings.

[0061] 11 is a flowchart of the new tag addition process. Next, the flow of the new tag addition process will be described with reference to FIG.

[0062] The tag editing unit 19 receives an input of the name of an item to which a new tag is to be added (step S101).

[0063] Next, the tag editing unit 19 receives input of information about the tag to be added (step S102).

[0064] Next, the tag editing unit 19 determines whether the tag to be added is a new tag or not based on the input tag information (step S103).

[0065] If the tag to be added is a new tag (step S103: Yes), the tag editing unit 19 newly registers tag data of the tag having the new tag word and stores it in the tag item storage unit 11 (step S104).

[0066] Next, the tag editing unit 19 newly registers a tag item corresponding to the newly created tag data and associates it with the specified item (step S105).

[0067] On the other hand, if the tag to be added is an existing tag (step S103: No), the tag editing unit 19 identifies the corresponding tag data from the tag data stored in the tag item storage unit 11. Then, the tag editing unit 19 increases the number of registrations of the identified tag data by one and rewrites the update date and time, thereby updating the tag data (step S106).

[0068] Next, the tag editing unit 19 identifies a tag item corresponding to the relevant tag from the tag items stored in the tag item storage unit 11. Then, the tag editing unit 19 adds information about the new item that has been input to the tag item (step S107).

[0069] Fig. 12 is a flowchart of a process at the time of a search according to the embodiment 1. Next, a flow of a process at the time of a search by the data search device 1 according to the embodiment 1 will be described with reference to Fig. 12. Here, a process after the tag matrix display screen is displayed by the tag matrix output unit 15 will be described.

[0070] The lower classification output unit 16 sets the hierarchy of the tag to be operated as the Nth tag, and initializes N to N=3 (step S201).

[0071] Next, the sub-classification output unit 16 receives an input of a selection made by performing a second operation on a cell that is an intersection of a set of tags in the tag matrix (step S202).

[0072] Next, the lower classification output unit 16 acquires the combination of the Nth tag in the selection target (step S203).

[0073] Next, the lower classification output unit 16 refers to the tag items stored in the tag item storage unit 11 and counts up the number of items for each tag of the Nth tag (step S204).

[0074] Next, the sub-classification output unit 16 adjusts the display width of each Nth tag in the columnar graph that displays the Nth tag in accordance with the number of items (step S205).

[0075] Next, the sub-classification output unit 16 displays a cylindrical graph representing the Nth tag on the tag matrix display screen (step S206).

[0076] Next, the lower classification output unit 16 receives input of the second operation for the tag group to be selected, and receives selection of the tag group to be selected (step S207).

[0077] Next, the sub-classification output unit 16 determines whether or not the list display button has been pressed (step S208). If the list display button has been pressed (step S208: Yes), the sub-classification output unit 16 proceeds to step S211.

[0078] On the other hand, if the list display button is not pressed (step S208: No), the lower-classification output unit 16 determines whether or not a tag in the next hierarchy exists (step S209).

[0079] If a tag of the next layer exists (step S209: Yes), the lower-classification output unit 16 increments N by 1 (step S210). After that, the lower-classification output unit 16 returns to step S203.

[0080] On the other hand, if there is no tag in the next layer (step S209: No), the lower-classification output unit 16 proceeds to step S211.

[0081] Thereafter, the lower classification output unit 16 displays the items belonging to the selected tag group on the item display screen (step S211).

[0082] As described above, the data search device according to this embodiment displays a tag matrix representing information on a combination of two tags. Furthermore, when a specific cell is selected in the displayed tag matrix, the data search device displays a cylindrical graph in which each item included in the cell is classified by other tags and the size of the area is adjusted according to the number of items. By using the tag matrix, a user can identify a group containing an object tagged with a desired tag. Furthermore, by selecting a specified group, a highly visible graph can be displayed showing the number of items included in the group classified by other tags. In other words, the size of the display area allows the user to easily grasp the approximate number of items in each tag group. Therefore, the user can easily find the desired search target during a search. Example 2

[0083] Next, a second embodiment will be described. The data search device according to this embodiment differs from the first embodiment in that, when displaying classifications of tags at a lower level than the tag matrix, the tags are displayed in different colors according to the number of accesses. The data search device according to this embodiment is also represented by the block diagram of Fig. 1. In the following description, the functions of each unit similar to those of the first embodiment may be omitted.

[0084] The subclassification output unit 16 receives input of a second operation for a cell on the tag matrix. Then, the subclassification output unit 16 acquires a combination of third tags for the selected cell. Next, the subclassification output unit 16 divides the items included in the cell for each acquired third tag. Next, the subclassification output unit 16 counts the number of items for each third tag. Then, the subclassification output unit 16 adjusts the display width of the tag group for each third tag in the columnar graph to match the number of items.

[0085] Next, for each item included in the selected cell, the subclassification output unit 16 inquires of the item output unit 17 about the most recent number of times the item has been accessed. Then, the subclassification output unit 16 checks the number of times each item included in the selected cell has been accessed.

[0086] Next, the sub-classification output unit 16 calculates the total number of most recent accesses of the items included in each third tag, and then determines the color scheme for each tag group so that the color of the display area becomes darker in order of the tag group with the largest total number of most recent accesses.

[0087] Thereafter, the sub-classification output unit 16 displays a cylindrical graph representing the tag group on the tag matrix display screen, with the width of the region varying according to the number of items and the color of the region varying according to the number of accesses. Fig. 13 is a diagram showing a display example of the tag matrix display screen displaying the classification of the third tag and subsequent tags according to the second embodiment.

[0088] For example, when the second operation is performed on the cell at the intersection of "Portal Site" and "Sales Promotion" on the tag matrix display screen 150 shown in FIG. 13, the sub-classification output unit 16 displays a cylindrical graph 151. The display width of the areas 501 to 504 representing the tag groups of the third tags #1 to #4 decreases as the number of items decreases. The colors of the areas 503, 502, 501, and 504 become lighter in this order. That is, the user can grasp the change in the number of accesses by looking at the colors. Therefore, the user can select the tag group of the third tag #3 in the area 503, which has the darkest color, assuming that a tag with a large number of recent accesses contains an item that has recently attracted attention. Furthermore, the user can easily select a tag group that is likely to contain the search target they are looking for by considering the display width and color.

[0089] Furthermore, when a specific tag group in the cylindrical graph representing the classification of the third tag is selected by the second operation, the subclassification output unit 16 classifies the items included in the selected tag group by the fourth tag. Then, the subclassification output unit 16 displays a cylindrical graph on the tag matrix display screen, with the display width adjusted based on the number of items and the color scheme adjusted based on the most recent access count, as in the case of the third tag.

[0090] For example, when the second operation is performed on the tag group of the third tag #2 in the cylindrical graph 151 on the tag matrix display screen 150 of Fig. 13, the sub-classification output unit 16 displays a cylindrical graph 152 on the tag matrix display screen 150. By checking the width and color of each area in the cylindrical graph 152, the user can grasp the status of the number of items in each tag group and the most recent access status.

[0091] The item output unit 17 stores the number of accesses for each item. For example, the item output unit 17 stores the number of accesses by incrementing the number of accesses by one each time an item is output to the output device 3. Furthermore, the item output unit 17 stores the time when each item was accessed, and subtracts the number of accesses before a predetermined time from the number of accesses for each item. Then, when the item output unit 17 receives an inquiry about the number of accesses for an item from the subclassification output unit 16, it notifies the subclassification output unit 16 of the number of accesses for the specified item.

[0092] In this embodiment, the item output unit 17 stores the number of accesses for each individual access, and the subclassification output unit 16 obtains the access count information from the item output unit 17 to determine the access count for each tag group. However, the method of obtaining the access count is not limited to this. For example, the subclassification output unit 16 may obtain the access count for each tag group using the number of views registered in the tag data stored in the tag item storage unit 11. Furthermore, when the tag matrix output unit 15 or the subclassification output unit 16 selects an item to be displayed, the tag matrix output unit 15 or the subclassification output unit 16 may register the user information of the user who requested the display and the access count of the displayed item in a display log registered in the display log storage unit 18. In this case, the subclassification output unit 16 can obtain the access count for each tag group by checking the display log stored in the display log storage unit 18.

[0093] Fig. 14 is a flowchart of a process at the time of a search according to the embodiment 2. Next, a flow of a process at the time of a search by the data search device 1 according to the embodiment 2 will be described with reference to Fig. 14. Here, a process after the tag matrix display screen is displayed by the tag matrix output unit 15 will be described.

[0094] The lower classification output unit 16 sets the hierarchy of the tag to be operated as the Nth tag, and initializes N to N=3 (step S301).

[0095] Next, the sub-classification output unit 16 receives an input of a selection made by performing a second operation on a cell that is an intersection of a set of tags in the tag matrix (step S302).

[0096] Next, the lower classification output unit 16 acquires the combination of the Nth tag in the selection target (step S303).

[0097] Next, the lower classification output unit 16 refers to the tag items stored in the tag item storage unit 11 and counts up the number of items for each tag of the Nth tag (step S304).

[0098] Next, the sub-classification output unit 16 adjusts the display width of each Nth tag in the columnar graph that displays the Nth tag in accordance with the number of items (step S305).

[0099] Next, the lower classification output unit 16 inquires of the item output unit 17 about the number of accesses to each item for each tag group of the Nth tag, and confirms the most recent number of accesses to each item (step S306).

[0100] Next, the subclassification output unit 16 calculates the number of accesses for each tag group by adding up the most recent access counts for each item in each tag group of the Nth tag. Then, the subclassification output unit 16 determines the color scheme so that the color of the display area becomes darker for tag groups with the most frequent accesses (step S307).

[0101] Next, the sub-classification output unit 16 displays a color-coded columnar graph representing the Nth tag on the tag matrix display screen (step S308).

[0102] Next, the lower classification output unit 16 receives input of the second operation for the tag group to be selected, and receives selection of the tag group to be selected (step S309).

[0103] Next, the sub-classification output unit 16 determines whether or not the list display button has been pressed (step S310). If the list display button has been pressed (step S310: Yes), the sub-classification output unit 16 proceeds to step S313.

[0104] On the other hand, if the list display button is not pressed (step S310: No), the lower-classification output unit 16 determines whether or not a tag in the next hierarchy exists (step S311).

[0105] If a tag of the next layer exists (step S311: Yes), the lower-classification output unit 16 increments N by 1 (step S312). After that, the lower-classification output unit 16 returns to step S303.

[0106] On the other hand, if there is no tag in the next layer (step S311: No), the lower-classification output unit 16 proceeds to step S313.

[0107] Thereafter, the lower classification output unit 16 displays the items belonging to the selected tag group on the item display screen (step S313).

[0108] As described above, when a specific cell is selected in the displayed tag matrix, the data search device according to this embodiment classifies each item contained in that cell by other tags. The data search device then displays the results of classifying each item contained in the cell by other tags using a cylindrical graph in which the size of the area is adjusted according to the number of items and the color is changed according to the number of accesses. By checking the size and color of the tag group, the user can grasp the general trend of the number of items contained and the most recent number of accesses. This allows the user to easily find the desired search target during a search.

[0109] In this embodiment, the color scheme is changed depending on the number of accesses, but other information may be used as the basis for determining the color scheme. For example, the color may be changed depending on the total access time rather than just the number of accesses. The number of accesses may be the actual number of users or the total number of users. Furthermore, the display log stored in the display log storage unit 18 may be used to obtain information about users who accessed each item, and the organizational scope of the users who accessed each item may be determined based on the department to which the users belong, and the color scheme may be determined for each tag group depending on the organizational scope. Alternatively, item ratings may be input, and the average rating may be calculated for each tag group, and the color scheme may be determined based on the rating.

[0110] In addition, although in this embodiment the state of each tag group is represented by color, other methods may be used as long as the user can recognize the difference. For example, each tag group may have a different pattern instead of a different color, or the volume or type of sound when the pointer is placed over it may be different. Example 3

[0111] Next, a third embodiment will be described. The data search device according to this embodiment differs from the first embodiment in that the classification of lower level tags is expressed using a triangle. The data search device according to this embodiment is also represented by the block diagram of Fig. 1. In the following description, the description of the functions of each unit similar to those of the first embodiment may be omitted.

[0112] 15 is a diagram showing an example of a tag matrix display screen displaying classifications of the third tag and subsequent tags according to Example 3. The subclassification output unit 16 according to this example displays classifications of tag groups of the third tag and subsequent tags as triangles with their bases at the bottom and vertices at the top on the tab matrix display screen. In this case, the subclassification output unit 16 displays tag groups from the bottom of the triangle in descending order of the number of items they contain in display areas of equal length in the vertical direction of the screen. This allows the subclassification output unit 16 to increase the display area for tag groups with a large number of items.

[0113] For example, a case will be described in which the cell at the intersection of "Portal Site" and "Sales Promotion" on the tag matrix display screen 106 in Fig. 15 is selected by the second operation. In this case, the sub-classification output unit 16 displays a triangle 161 on the tag matrix display screen 106, which indicates the classification results by third tag for the items included in the cell. In the triangle 161, the third tag #1 with the largest number of items is placed at the bottom of the screen, the third tag #2 with the next largest number of items is placed above it, the third tag #3 with the next largest number of items is placed above it, and the third tag #4 with the smallest number of items is placed at the top.

[0114] Furthermore, when the tag group of the third tag #2 in the triangle 161 is selected, the sub-classification output unit 16 displays a triangle 162 indicating the classification results of the items included in that tag group with the fourth tag. In the triangle 162, the tag groups of the fourth tag are also arranged from the bottom in descending order of the number of items.

[0115] As explained above, the data search device according to this embodiment uses triangles to represent the difference in the number of items included in each tag group in the classification from the third tag onwards. In this way, any representation method other than cylindrical graphs that can vary the size of the display area for each tag group can improve user visibility, allowing users to easily find the desired search target during a search. Example 4

[0116] Next, a fourth embodiment will be described. The data search device according to this embodiment differs from the first embodiment in that the classification of lower level tags is expressed using a pie chart. The data search device according to this embodiment is also represented by the block diagram of Fig. 1. In the following description, the description of the functions of each unit similar to those of the first embodiment may be omitted.

[0117] 16 is a diagram showing a display example of a tag matrix display screen displaying classifications from the third tag onwards according to Example 4. The subclassification output unit 16 according to this example displays classifications of tag groups from the third tag onwards in terms of the size of the pie chart on the tab matrix display screen. In this case, the subclassification output unit 16 displays the classifications of tag groups from the third tag onwards in descending order of the number of items included, in a display area corresponding to the percentage of the whole, going clockwise from the apex of the pie chart (12 o'clock on a clock: from here onwards it is expressed in terms of clock time). This allows the subclassification output unit 16 to increase the display area of tag groups with a large number of items.

[0118] For example, a case will be described in which the cell at the intersection of "Portal Site" and "Sales Promotion" on the tag matrix display screen 107 in Fig. 16 is selected by the second operation. In this case, the sub-classification output unit 16 displays a pie chart 171 on the tag matrix display screen 170, showing the classification results by third tag for the items included in the cell. In the pie chart 171, the third tag #1 with the largest number of items is arranged from 12 o'clock to 3 o'clock on the screen, the third tag #2 with the next largest number of items is arranged in the next position clockwise, the third tag #3 with the next largest number of items is arranged in the next position further clockwise, and the third tag #4 with the smallest number of items is arranged from 9 o'clock to 12 o'clock.

[0119] Furthermore, when a tag group of the third tag #2 in the pie chart 171 is selected, the sub-classification output unit 16 displays a pie chart 172 showing the classification results of the items included in that tag group by the fourth tag. In the pie chart 172, each tag group of the fourth tag is arranged clockwise from the 12 o'clock position in descending order of the number of items.

[0120] As described above, the data search device according to this embodiment uses a pie chart to represent the difference in the number of items included in each tag group in the classification from the third tag onwards. In this way, any representation method other than a cylindrical graph that can vary the size of the display area for each tag group can improve user visibility, allowing the user to easily find the desired search target during a search.

[0121] Next, an example of the hardware configuration of a computer that realizes the same functions as the data search device 1 shown in the above embodiment will be described. Fig. 17 is a diagram showing an example of the hardware configuration of a computer that realizes the same functions as the data search device.

[0122] As shown in FIG. 17, a computer 90 has a CPU (Central Processing Unit) 91 that executes various types of arithmetic processing, a RAM (Random Access Memory) 92 that is a main storage device that temporarily stores various types of information, and a hard disk drive 93 that is an auxiliary storage device. The computer 90 also has a communication device 94 that transmits and receives data to and from external devices, etc., via a wired or wireless network, and an interface device 95. The CPU 91, RAM 92, hard disk drive 93, communication device 94, and interface device 95 are connected to a bus 96.

[0123] The hard disk drive 93 can realize the functions of the tag item storage unit 11, tag matrix storage unit 13, and display log storage unit 18 illustrated in Fig. 1. The hard disk drive 93 also stores various programs including programs for realizing the functions of the tag matrix editing unit 12, item number tallying unit 14, tag matrix output unit 15, sub-classification output unit 16, item output unit 17, and tag editing unit 19 illustrated in Fig. 1.

[0124] The CPU 91 reads various programs stored in the hard disk drive 93, expands them on the RAM 92, and generates and executes processes. As a result, the CPU 91 realizes the functions of the tag matrix editing unit 12, the item number tallying unit 14, the tag matrix output unit 15, the subclassification output unit 16, the item output unit 17, and the tag editing unit 19 illustrated in FIG.

[0125] 1 do not necessarily have to be stored in the hard disk drive 93 from the beginning. For example, each program may be stored in a "portable physical medium" such as a flexible disk (FD), CD-ROM, DVD, magneto-optical disk, or IC card that is inserted into the computer 90. Then, the CPU 91 may read and execute each program from the portable physical medium using the interface device 95 or the like. [Explanation of symbols]

[0126] 1 Data search device 2 Input devices 3 Output Devices 4 Database 11 Tag item storage section 12 Tag Matrix Editorial Department 13 Tag matrix storage unit 14 Item counting section 15 Tag matrix output section 16 Subclassification output section 17 Item output section 18 Display log storage unit 19 Tag Editorial Department

Claims

1. A data search device that searches for data stored in a database using row tags and column tags, which are character strings, a tag item storage unit that stores a correspondence between the row tags and the column tags and the data to which the row tags and the column tags are added; a tag matrix storage unit that stores a tag matrix having elements indicated by the row tags and the column tags, the tag matrix displaying information about the data to which the corresponding row tags and column tags are attached for each of the elements; a tag matrix output unit that displays the tag matrix on a display device; a subclassification output unit that, when a specific element to which a first row tag and a first column tag are assigned and which corresponds to first data is selected from the elements included in the tag matrix displayed on the display device, identifies a predetermined number of second tags that have the highest total number assigned to the first data from among tags other than the first row tag and the first column tag, performs a first classification of the first data according to characteristics of each of the first data using the predetermined number of second tags, and generates a plurality of first groups, and, when a specific first group is designated from the generated plurality of first groups by a predetermined operation, arranges a list of items that are the first data to which the second tags included in the designated specific first group are assigned near the plurality of first groups and displays it on the display device; A data search device comprising:

2. 2. The data search device according to claim 1, wherein the subclassification output unit changes the state of a display area according to the feature amount of each of the first groups, and displays each of the first groups on the display device.

3. 3. The data search device according to claim 2, wherein the lower classification output unit changes the size of the display area in accordance with the number of the first data items belonging to each of the first groups.

4. 4. The data search device according to claim 3, wherein the subclassification output unit displays each of the first groups in a stacked manner using a cylindrical graph, and changes the width of layers in the cylindrical graph depending on the number of the first data belonging to each of the first groups.

5. 4. The data search device according to claim 3, wherein the subclassification output unit displays each of the first groups stacked inside a triangle, and arranges the first data items in descending order of the number of first data items belonging to each of the first groups from the bottom side of the triangle.

6. A data search device described in any one of claims 2 to 5, characterized in that the sub-classification output unit changes the color scheme of the display area of each of the first groups based on the number of accesses of the first data belonging to each of the first groups.

7. The data search device according to any one of claims 1 to 6, characterized in that the sub-classification output unit, upon receiving a selection of one of the first groups, performs a second classification of the first data belonging to the selected first group using the first row tag, the first column tag, and a third tag other than the first row tag, the first column tag, and the second tag corresponding to the selected first group, to generate a plurality of second groups, and changes the size of a display area according to the number of second data belonging to each of the second groups to display information of each of the second groups on the display device.

8. A data search method for searching data stored in a database using row tags and column tags, which are character strings. storing a correspondence relationship between the row tags and the column tags, the row tags and the column tags, and the data to which the row tags and the column tags are added; displaying, on a display device, a tag matrix having elements indicated by the row tags and the column tags, the tag matrix displaying information about the data to which the corresponding row tags and column tags are attached for each of the elements; When a specific element to which a first row tag and a first column tag are assigned and which corresponds to first data is selected from the elements included in the tag matrix displayed on the display device, a predetermined number of second tags with the highest total number assigned to the first data are identified from among tags other than the first row tag and the first column tag, the first data is classified according to the characteristics of each of the first data using the predetermined number of second tags to generate a plurality of first groups, and when a specific first group is designated from the generated plurality of first groups by a predetermined operation, a list of items which are the first data to which the second tags included in the designated specific first group are assigned is arranged near the plurality of first groups and displayed on the display device. A data search method characterized in that the processing is executed by a computer.

9. A data search program that causes a computer to execute a process of searching data stored in a database using row tags and column tags, which are character strings. storing a correspondence relationship between the row tags and the column tags, the row tags and the column tags, and the data to which the row tags and the column tags are added; displaying, on a display device, a tag matrix having elements indicated by the row tags and the column tags, the tag matrix displaying information about the data to which the corresponding row tags and column tags are attached for each of the elements; When a specific element to which a first row tag and a first column tag are assigned and which corresponds to first data is selected from the elements included in the tag matrix displayed on the display device, a predetermined number of second tags with the highest total number assigned to the first data are identified from among tags other than the first row tag and the first column tag, the first data is classified according to the characteristics of each of the first data using the predetermined number of second tags to generate a plurality of first groups, and when a specific first group is designated from the generated plurality of first groups by a predetermined operation, a list of items which are the first data to which the second tags included in the designated specific first group are assigned is arranged near the plurality of first groups and displayed on the display device. A data retrieval program that causes the computer to execute a process.

Images