Retrieval menu creation device, a retrieval menu creation method, and a recording medium that stores a retrieval menu creation program

• US 6,219,665 B1
• Filed: 06/05/1998
• Issued: 04/17/2001
• Est. Priority Date: 06/06/1997
• Status: Expired due to Term

• Current Assignee: Panasonic Intellectual Property Corporation of America (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Shiomi, Takakazu
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): ROBINSON, GRETA LEE
• Application Number:
  US09/092,172
• Time in Patent Office: 1,047 Days
• Field of Search: 707/6, 707/2, 707/100, 707/45, 707/1, 707/102, 345/352, 345/354, 345/146
• US Class Current: 707/750
• CPC Class Codes: G06F 16/3323 using document space presentation or visualization, e.g. category, hierarchy or range presentation and selection
• CPC Class Codes: Y10S 707/954 Relational
• CPC Class Codes: Y10S 707/956 Hierarchical
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99936 Pattern matching access

Abstract

Data storage unit 101 stores input data. Data feature extraction unit 102 extracts features from the input data, and writes the features in data storage unit 101. Categorization pattern storage unit 103 stores categorization pattern information 700. Relating unit 105 relates the input data to items included in the categorization patterns using the features. First categorization pattern selection unit 106 selects the most suitable categorization pattern for creating a retrieval menu. Menu creation unit 107 creates the retrieval menu based on the selected categorization pattern. Display unit 108 displays the created retrieval menu.

Claims

1. A retrieval menu creation device that creates retrieval menus based on a plurality of pieces of input data comprising:

a reception means for receiving a plurality of pieces of input data;

a data storage means for storing the plurality of pieces of input data;

a writing means for writing the received plurality of pieces of input data into the data storage means;

a categorization pattern storage means for storing a plurality of categorization patterns each of which includes a plurality of items

a relating means for relating pieces of input data that are stored in the storage means to corresponding items in the categorization patterns;

a categorization pattern selection means for selecting categorization patterns from the plurality of categorization patterns that are stored in the categorization pattern storage means based on a distribution of the input data related to the plurality of categorization patterns;

a menu extraction means for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices; and

a display means for displaying the retrieval menu that is created by the menu extraction means.

2. The retrieval menu creation device according to claim 1 wherein:

the data storage means stores pieces of data identification data for identifying the plurality of pieces of input data, the input data represents document data and

the relating means includes;

a feature extraction unit including a data reading element for reading the plurality of pieces of document data from the data storage means, and for extracting features from the read pieces of document data;

a morphological analysis element for performing a morphological analysis on the read pieces of document data, and for dividing the read pieces of document data into morphemes; and

an important word extraction element for extracting important words from the morphemes, and for setting the extracted important words as features;

an item retrieval unit for retrieving items that correspond to extracted features from the categorization pattern storage means; and

a data relating unit for adding data identification data for read pieces of document data to the retrieved items.

3. The retrieval menu creation device according to claim 2 wherein the important word extraction element includes:

an important word candidate extraction element for extracting important word candidates from the morphemes;

a frequency-in-use count element for counting frequency-in-use of each of the extracted important word candidates in a piece of document data; and

a predetermined frequency extraction element for extracting each important word candidate whose frequency-in-use in a piece of document data is greater than a predetermined value from the extracted important word candidates as an important word, and for setting the extracted important words as features.

4. The retrieval menu creation device according to claim 2 wherein the important word extraction element includes:

an important word candidate extraction element for extracting important word candidates from the morphemes;

a frequency-in-use count element for counting frequency-in-use of each of the extracted important word candidates in a piece of document data; and

a predetermined number extraction element for extracting a first predetermined number of important word candidates as important words from the extracted important word candidates in order of decreasing frequency-in-use in a piece of document data, and for setting the important words as features.

5. The retrieval menu creation device according to claim 1 wherein:

the data storage means stores pieces of data identification data as a relational database for identifying the plurality of pieces of input data wherein one of the plurality of pieces of input data represents one row in the relational database and includes at least one value list;

the relating means includes;

a feature extraction unit for reading the plurality of pieces of input data from the data storage means, and for extracting features from the read pieces of input data;

an item retrieval unit for retrieving items that correspond to extracted features from the categorization pattern storage means; and

a data relating unit for adding data identification data for read pieces of input data to the retrieved items.

6. The retrieval menu creation device according to claim 1 wherein:

the data storage means stores pieces of data identification data for identifying the plurality of pieces of input data; and

the relating means includes;

a feature extraction unit for reading the plurality of pieces of input data from the data storage means, and for extracting features from the read pieces of input data;

an item retrieval unit for retrieving items that correspond to extracted features from the categorization pattern storage means; and

a data relating unit for adding data identification data for read pieces of input data to the retrieved items to relate the read piece of input data that includes the extracted features to the retrieved items; and

the categorization pattern storage means includes the plurality of items included in a categorization pattern in a hierarchy.

7. The retrieval menu creation device according to claim 6 further comprising:

an input reception means for receiving choices that have been selected by a user from the choices in the retrieval menu;

a judging means for judging, when the input reception means receives the selected choices, whether another retrieval menu should be created or whether input data should be displayed based on a second predetermined standard;

an item menu creation means for extracting, when the judging means judges that another retrieval menu should be created, items to which input data has been related from items below items corresponding to the selected choices in the hierarchy, and for creating another retrieval menu in which the extracted items represent choices;

an input data menu display means for displaying, when the judging means judges that another retrieval menu should be created, the created other retrieval menu, and for reading, when the judging means judges that input data should be displayed, input data that has been related to any of the items corresponding to the selected choices and the items below the corresponding item in the hierarchy from the data storage means, and for displaying the read input data; and

a repetition control means for controlling the input reception means, the judging means, the item menu creation means, and the input data menu display means so that a process in which the input reception means receives choices, the judging means judges, the item menu creation means creates another retrieval menu, and the input data menu display means displays the created other retrieval menu should be repeated until the input data menu display means displays the read input data.

8. The retrieval menu creation device according to claim 7 wherein the second predetermined standard used by the judging means represents a standard by which, when a number of pieces of the input data that have been related to any of the items corresponding to the selected choices and the items below the corresponding items in the hierarchy is smaller than or equal to a predetermined value, the input data is displayed, while when the number of pieces of the input data that have been related to any of the items corresponding to the selected choices and the items below the corresponding items in the hierarchy is larger than the predetermined value, another retrieval menu is created.

9. The retrieval menu creation device according to claim 1 wherein the categorization pattern selection means includes:

an item number count unit for counting a number of items related to input data for each of the plurality of categorization patterns; and

an items number selection unit for selecting categorization patterns in which the number of items related to input data is close to a second predetermined number.

10. The retrieval menu creation device according to claim 9 further comprising:

a priority ranking storage means for storing a priority ranking for each of the plurality of categorization patterns; and

a priority ranking selection means for reading, when the item number selection unit selects more than once categorization pattern, a priority ranking for each of the selected categorization patterns, and for selecting categorization patterns using the read priority rankings.

13. The retrieval menu creation device according to claim 1 wherein the categorization pattern selection means includes:

a variance calculation unit for calculating a variance in a number of pieces of input data that are related to each item in a categorization pattern, for each categorization pattern; and

a variance selection unit for selecting categorization patterns that have the smallest calculated variance of the number of pieces of the input data that has been related to items.

14. The retrieval menu creation device according to claim 13 further comprising:

a priority ranking storage means for storing a priority ranking for each of the plurality of categorization patterns; and

a priority ranking selection means for reading, when the variance selection unit selects more than one categorization pattern, a priority ranking for each of the selected categorization patterns, and for selecting categorization patterns using the read priority rankings.

15. The retrieval menu creation device according to claim 1 wherein the categorization pattern selection means includes:

a priority ranking unit for storing a priority ranking for each of the plurality of categorization patterns; and

a priority ranking selection unit for reading a priority ranking for each of the plurality of categorization patterns, and for selecting categorization patterns using the read priority rankings.

16. The retrieval menu creation device according to claim 1 further comprising:

a switch reception means for receiving a user indication switch from a retrieval menu that is displayed on the display means to another retrieval menu; and

a second categorization pattern selection means for selecting, when the switch reception means receives the user indication to switch from a retrieval menu that is displayed on the display means to another retrieval menu, categorization patterns from categorization patterns apart from the categorization patterns that were selected by the categorization pattern selection means, based on the first predetermined standard that was used by the categorization pattern selection means, wherein

the menu extraction means creates another retrieval menu from the categorization patterns that have been selected by the second categorization pattern selection means.

17. The retrieval menu creation device according to claim 1 further comprising:

an input reception means for receiving choices that have been selected by a user from the choices in the retrieval menu, wherein

the relating unit further includes;

a selection input data extraction means for extracting each piece of input data that has been related to items corresponding to the selected choices as selection data;

a relation cancel means for canceling relations between each extracted piece of input data and items; and

a selection input data relating means for relating the selection data to items included in categorization patterns apart from categorization patterns that include the item corresponding to the selected choice,

wherein the categorization pattern selection means selects categorization patterns that include items to which the selection data was related by the selection input data relating means, based on the first predetermined standard,

wherein the menu extraction means creates another retrieval menu from the selected categorization patterns, and

wherein the display means displays the other retrieval menu that has been created by the menu extraction means.

18. The retrieval menu creation device according to claim 1 further comprising:

a categorization pattern input reception means for receiving a categorization pattern and instructions to edit the categorization pattern from the user, wherein instructions to edit a categorization pattern include instructions to register the categorization pattern and instructions to delete the categorization pattern; and

a categorization pattern editing means for editing the plurality of categorization patterns that are stored in the categorization pattern storage means based on the categorization pattern and the instructions to edit the categorization pattern that have been received by the categorization pattern input reception means.

11. The retrieval menu creation device that creates retrieval menus based on a plurality of pieces of input data comprising:

a reception means for receiving a plurality of pieces of input data;

a data storage means for storing the plurality of pieces of input data;

a writing means for writing the received plurality of pieces of input data into the data storage means;

a categorization pattern storage means for storing a plurality of categorization patterns each of which includes a plurality of items

a relating means for relating pieces of input data that are stored in the data storage means to corresponding items in the categorization patterns;

a categorization pattern selection means for selecting categorization patterns from the plurality of categorization patterns that are stored in the categorization pattern storage means based on a distribution of the input data related to the plurality of categorization patterns, wherein in the categorization pattern selection means includes;

an input data type number count unit for counting a number of types of input data that have been related to items for each of the plurality of categorization patterns; and

an input data type number selection unit for selecting categorization patterns in which a largest number of types of input data have been related to items; and

a menu extraction means for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices; and

a display means for displaying the retrieval menu that is created by the menu extraction means.

12. The retrieval menu creation device according to claim 11 further comprising:

a priority ranking storage means for storing a priority ranking for each of the plurality of categorization patterns; and

a priority ranking selection means for reading, when the input data type number selection unit selects more than one categorization pattern, a priority ranking for each of the selected categorization patterns, and for selecting categorization patterns using the read priority rankings.

19. A retrieval menu creation method that is used by a retrieval menu creation device which comprises a data storage means for storing a plurality of pieces of input data, and a categorization pattern storage means for storing a plurality of categorization pattern which each include a plurality of items, wherein no item that is included in a categorization pattern is related to any item that is included in another categorization pattern, and creates retrieval menus based on the plurality of pieces of input data the method comprising:

a receiving step for receiving a plurality of pieces of input data;

a relating step for relating pieces of input data that are stored in the data storage mans to corresponding items;

a writing step for writing the received plurality of pieces of input data into the data storage means;

a categorization pattern selection step for selecting categorization patterns from the plurality of categorization patterns that are stored in the categorization pattern storage means based on a distribution of the input data related to the plurality of categorization patterns;

a menu extraction step for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices; and

a display step for displaying the retrieval menu that is created by the menu extraction step.

20. The retrieval menu creation method according to claim 19 wherein the data storage means stores pieces of data identification data for identifying the plurality of pieces of input data;

and the relating step includes;

a feature extraction step for reading the plurality of pieces of input data from the data storage means, and for extracting features from the read pieces of input data;

an item retrieval step for retrieving items that correspond to extracted features from the categorization pattern storage means; and

a data relating step for adding data identification data for read pieces of input data to the retrieved items, and wherein the plurality of pieces of input data stored in the data storage means represents a plurality of pieces of document data; and

the feature extraction step includes;

a data step for reading the plurality of pieces of document data from the data storage means;

a morphological analysis step for performing a morphological analysis on the read pieces of document data, and for dividing the read pieces of document data into morphemes; and

an important word extraction step for extracting important words from the morphemes, and for setting the extracted important words as features.

21. The retrieval menu creation method according to claim 20 wherein the important word extraction step includes:

an important word candidate extraction step for extracting important word candidates from the morphemes;

a frequency-in-use count step for counting frequency-in-use of each of the extracted important word candidates in a piece of document data; and

a predetermined frequency extraction step for extracting each important word candidate whose frequency-in-use in a piece of document data is greater than a predetermined value from the extracted important word candidates as an important word, and for setting the extracted important words as features.

22. The retrieval menu creation method according to claim 20 wherein the important word extraction step includes:

an important word candidate extraction step for extracting important word candidates from the morphemes;

a frequency-in-use count step for counting frequency-in-use of each of the extracted important word candidates in a piece of document data; and

a predetermined number extraction step for extracting a first predetermined number of important word candidates as important words from the extracted important word candidates in order of decreasing frequency-in-use in a piece of document data, and for setting the important words as features.

23. The retrieval menu creation method according to claim 19, the data storage means stores the plurality of pieces of input data as a relational database, with one of the plurality of pieces of input data represents one row in the relational database and includes at least one value list;

and the relating step includes;

a feature extraction step for reading the plurality of pieces of input data from the data storage means, and for extracting features from the read pieces of input data;

an item retrieval step for retrieving items that correspond to extracted features from the categorization pattern storage means; and

a data relating step for adding data identification data for read pieces of input data to the retrieved items, and wherein the feature extraction step includes;

a data reading step for reading the plurality of pieces of input data from the data storage means; and

a data extraction for extracting values from the read piece of input data as features.

24. The retrieval menu creation method according to claim 19 wherein the categorization pattern storage means includes the plurality of items included in a categorization pattern in a hierarchy, wherein the relating step includes:

a feature extraction step for reading the plurality of pieces of input data from the data storage means, and for extracting features from the read pieces of input data;

an item retrieval step for retrieving items that corresponding to extracted features from the categorization pattern storage means; and

a data relating step for adding data identification data for read pieces of input data to the retrieved items, and wherein

the relating step further includes an item retrieval step for retrieving the items that correspond to the features from the categorization pattern storage means, and

the data relating step relates the read piece of input data that includes the extracted features to the retrieved items.

25. The retrieval menu creation method according to claim 24 further comprising:

an input reception step for receiving choices that have been selected by a user from the choices in the retrieval menu;

a judging step for judging, when the input reception step receives the selected choices, whether another retrieval menu should be created or whether input data should be displayed based on a second predetermined standard;

an item menu creation step for extracting, when the judging step judges that another retrieval menu should be created, items to which input data has been related from items below items corresponding to the selected choices in the hierarchy, and for creating another retrieval menu in which the extracted items represent choices;

an input data menu display step for displaying, when the judging step judges that another retrieval menu should be created, the created other retrieval menu, and for reading, when the judging step judges that input data should be displayed, input data that has been related to any of the items corresponding to the selected choices and the items below the corresponding item in the hierarchy from the data storage means, and for displaying the read input data; and

a repetition control step for controlling the input reception step, the judging step, the item menu creation step, and the input data menu display step so that a process in which the input reception step receives choices, the judging step judges, the item menu creation step creates another retrieval menu, and the input data menu display step displays the created other retrieval menu should be repeated until the input data menu display step displays the read input data.

26. The retrieval menu creation method according to claim 25 wherein the second predetermined standard used by the judging step represents a standard by which, when a number of pieces of the input data that have been related to any of the items corresponding to the selected choices and the items below the corresponding items in the hierarchy is smaller than or equal to a predetermined value, the input data is displayed, while when the number of pieces of the input data that have been related to any of the items corresponding to the selected choices and the items below the corresponding items in the hierarchy is larger than the predetermined value, another retrieval menu is created.

27. The retrieval menu creation method according to claim 19 wherein the categorization pattern selection step includes:

an item number count step for counting a number of items related to input data for each of the plurality of categorization patterns; and

an item number selection step for selecting categorization patterns in which the number of items related to input data is close to a second predetermined number.

28. The retrieval menu creation method according to claim 27 wherein the retrieval menu creation device further comprises a priority ranking storage means for storing a priority ranking for each of the plurality of categorization patterns;

and the method further comprises a priority ranking selection step for reading, when the item number selection step selects more than one categorization pattern, a priority ranking for each of the selected categorization patterns, and for selecting categorization patterns using the read priority rankings.

29. The retrieval menu creation method according to claim 19 wherein the retrieval menu creation device further comprises a priority ranking storage means for storing a priority ranking for each of the plurality of categorization patterns;

and the method further comprises a priority ranking selection step for reading a priority ranking for each of the plurality of categorization patterns, and for selecting categorization patterns using the read priority rankings.

30. The retrieval menu creation method according to claim 19 further comprising:

an input reception step for receiving choices that have been selected by a user from the choices in the retrieval menu, wherein

the relating step further includes;

a selection input data extraction step for extracting each piece of input data that has been related to items corresponding to the selected choices as selection data;

a relation cancel step for canceling relations between each extracted piece of input data and items; and

a selection input data relating step for relating the selection data to items included in categorization patterns apart from categorization patterns that include the item corresponding to the selected choice,

wherein the categorization pattern selection step selects categorization patterns that include items to which the selection data was related by the selection input data relating step, based on the first predetermined standard,

wherein the menu extraction step creates another retrieval menu from the selected categorization patterns, and

wherein the display step displays the other retrieval menu that has been created by the menu extraction step.

31. A computer-readable storage medium for use by a computer which comprises a data storage means for storing a plurality of pieces of input data and a categorization pattern storage means for storing a plurality of categorization patterns which each include a plurality of items, wherein no item that is included in a categorization pattern is related to any item that is included in another categorization pattern, the storage medium storing a program in which retrieval menus are created based on the plurality of pieces of input data, the storage medium comprising:

a relating step for relating pieces of input data that are stored in the data storage means to corresponding items;

a categorization pattern selection step for selecting categorization patterns from the plurality of categorization patterns that are stored in the categorization pattern storage means based on a distribution of the input data related to the plurality of categorization patterns; and

a menu extraction step for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices.

32. A retrieval menu creation device that creates retrieval menus based on a plurality of pieces of input data comprising:

a reception means for receiving a plurality of pieces of input data;

a data storage means for storing the plurality of pieces of input data;

a writing means for writing the received plurality of pieces of input data into the data storage means;

a categorization pattern storage means for storing a plurality of categorization patterns each of which includes a plurality of items, wherein each item included in each one of the plurality of categorization patterns is exclusive to that categorization pattern;

a relating means for relating pieces of input data with items in the plurality of categorization patterns;

a categorization pattern selection means for selecting categorization patterns from the plurality of categorization patterns that are stored in the categorization pattern storage means based on a distribution of the input data related to the plurality of categorization patterns;

a menu extraction means for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices in the retrieval menu; and

a display means enabling a display of the retrieval menu that is created by the menu extraction means.

33. A method of creating retrieval menus based on input data, comprising the steps of:

receiving a plurality of pieces of input data;

storing the plurality of pieces of input data;

storing a plurality of categorization patterns each of which includes a plurality of items that are exclusive to at least one of the plurality of categorization patterns;

relating pieces of input data with items in the plurality of categorization patterns;

selecting categorization patterns from the plurality of categorization patterns that are stored based on a distribution of the input data related to the plurality of categorization patterns;

creating a retrieval menu by extracting items which relate to the piece of input data determined with the categorization patterns, the retrieval menu includes the extracted items as choices in the retrieval menu; and

displaying the retrieval menu.

34. A machine readable medium for storing executable data instructions that can create retrieval menus based on a plurality of pieces of input data, comprising:

writing data for writing the plurality of pieces of input data into a data storage memory;

a plurality of categorization pattern data, each categorization pattern includes a plurality of items, wherein no item that is included in each one of the categorization patterns is related to any item that is included in another categorization pattern;

relating data for relating pieces of input data that are stored in the storage memory to corresponding items in the categorization patterns;

a categorization pattern selection data for selecting categorization patterns from the plurality of categorization patterns based on a distribution of the input data related to the plurality of categorization patterns;

a menu extraction data for extracting items to which pieces of input data are related from the selected categorization patterns, and for creating a retrieval menu which includes the extracted items as choices; and

a display data for enabling the displaying of the retrieval menu that is created by the menu extraction data.