Information storage and retrieval system and display method therefor

• US 5,696,916 A
• Filed: 09/21/1992
• Issued: 12/09/1997
• Est. Priority Date: 03/27/1985
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Hashimoto, Tetsuya, Fujisawa, Hiromichi, Yamazaki, Naoko, Kiuchi, Itsuko
• Primary Examiner(s): Fetting, Anton W.
• Application Number:
  US07/947,536
• Time in Patent Office: 1,905 Days
• Field of Search: 395/12, 395/51, 395/54, 395/63, 395/155, 395/157, 395/160, 395/161, 395/326, 395/340, 395/356, 395/961, 395/615, 395/761, 364/419.19
• US Class Current: 715/853
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs

Abstract

An information storage and retrieval system which permits storage of fact information represented in terms of conceptual concepts expressing things and relational concepts expressing relations therebetween on a knowledge base, easy inputting and correction of the fact information, easy browsing and retrieval of the fact information and display of the fact information with effective use of a limited view area, and also to provide a display method used in the information storage and retrieval system. The system includes a device for inputting fact information with respect to a specific relational concept according to a template pattern by using a multi-window function on the basis of information obtained from a concept dictionary for defining hierarchical relations between concepts and a grammar rule for defining semantic structures of cases and concepts allowed to be used in the cases; a device for storage and retrieval of fact information; and a device for displaying fact information in a fact window and displaying a hierarchy of conceptual concepts and relational concepts in a node tree window.

Claims

1. An information storage and retrieval system comprising:

a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept, a numerical value conceptual concept and a unit conceptual concept, relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural conceptual concepts;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

means for inputting fact information represented in terms of said conceptual concepts, said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept, and defining a relation between a relational concept or an attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form, wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

a knowledge base for storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

means for generating a retrieval request for a conceptual concept based on said templates, said conceptual concepts and said stored fact information as a result of interaction with a user;

means for retrieving a requested conceptual concept based on said retrieval request; and

means for displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form for viewing by said user.

2. An information storage and retrieval system according to claim 1, wherein said relational concepts include verbal concepts representing verbs, and said attribute name concepts include attribute concepts representing attributes.

3. An information storage and retrieval system according to claim 2, wherein said conceptual concepts include noun concepts representing nouns, unit name concepts representing names of numerical values, and unit concepts representing units, and

wherein said system further comprises:

means for managing said relational concepts by is-a relations, each is-a relation defining a lower relational concept as a thing defined by an upper relational concept, and managing said conceptual concepts by part-whole relations, each part-whole relation defining a lower concept as a thing included within the whole thing defined by an upper concept.

4. An information storage and retrieval system according to claim 3, wherein said system further comprises:

a second knowledge base for storing a concept relation model for managing and expressing said verbal concepts and said attribute concepts in said relational concepts by is-a relations.

5. An information storage and retrieval system according to claim 3, wherein said system further comprises:

means for retrieving fact information stored in terms of said relational concepts and said attribute name concepts and at least one conceptual concept by using at least one upper relational concept or attribute name concept positioned in said hierarchical tree form over said relational concepts and said attribute name concepts.

6. An information storage and retrieval system according to claim 3, wherein said system further comprises:

means for storing said noun conceptual concepts and said unit conceptual concepts in terms of names each expressed by at least one string of characters, and managing said noun conceptual concepts and said unit conceptual concepts by using one of said strings as a representative name.

7. An information storage and retrieval system according to claim 2, wherein said system further comprises:

storing information of standard cases to define semantic information in said templates, and a semantic priority of a semantic order among standard cases necessary for representing said verbal concepts.

8. An information storage and retrieval system according to claim 2, wherein said system further comprises:

means for storing said relation concepts in terms of names each expressed by at least one string of characters, and managing said relational concepts by using one of said strings as a representative name.

9. An information storage and retrieval system according to claim 2, wherein said system further comprises:

means for storing said verbal concepts in terms of etyma of corresponding verbs each expressed by a string of characters, and storing information concerning strings of characters corresponding to conjugations thereof and a difference in kinds of conjugations.

10. An information storage and retrieval system according to claim 1, wherein template patterns are respectively defined for said templates, and

wherein said system further comprises:

means for expressing semantic information in a pseudo-natural language by using said template patterns.

11. An information storage and retrieval system according to claim 1, wherein said system further comprises:

means for selecting a template based on semantic information of said templates and said requested conceptual concept such that said requested conceptual concept corresponds to at least one case of said semantic information.

12. An information storage and retrieval system according to claim 1, wherein said system further comprises:

means for selecting said templates based on semantic information of said templates to selected templates corresponding to said relational concepts or said attribute name concepts.

13. An information storage and retrieval system comprising:

a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept a numerical value conceptual concept and a unit conceptual concept, relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

means for inputting fact information represented in terms of said conceptual concepts, said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept, and defining a relation between a relational concept or an attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

a knowledge base for storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

means for generating a retrieval request for a conceptual concept based on said templates, said conceptual concepts and said stored fact information as a result of interaction with a user;

means for retrieving a requested conceptual concept based on said retrieval request;

means for displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form for viewing by said user;

a second knowledge base for storing a concept relation model for expressing strings of characters for prepositions added before semantic information of said relational concepts and said attribute name concepts and postpositions added after semantic information of said relational concepts and said attribute name concepts so that said templates are able to be applied in a plurality of pseudo-natural languages by adding said prepositions or said postpositions; and

means for adding said prepositions and postpositions before and after said semantic information of said relational concepts and said attribute name concepts.

14. An information storage and retrieval system comprising:

a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept, a numerical value conceptual concept and a unit conceptual concept, relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

means for inputting fact information represented in terms of said conceptual concepts said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept, and defining a relation between a relational concept or an attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form, wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

means for inputting a plurality of conceptual concepts, adding general information to said conceptual concepts so that the expression of said fact information including said conceptual concepts to which said general information has been added becomes general, and adding indefinite information to said conceptual concepts so that the expression of said conceptual concepts to which said indefinite information has been added becomes indefinite;

means for quoting other fact information instead of said conceptual concept, so that said quoted other fact information becomes an objective expression;

a knowledge base for storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

means for managing said stored fact information based on said semantic information defined by said templates; and

means for displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form which includes at least one conceptual concept.

15. An information storage and retrieval system according to claim 14, wherein said relational concepts include verbal concepts representing verbs, and said attribute name concepts includes attribute concepts representing attributes;

and

wherein said system further comprises;

means for managing said conceptual concepts, said relational concepts said attribute name concepts by is-a relations, each is-a relation defining a lower concept as a thing defined by an upper concept, and managing said conceptual concepts by part-whole relations, each part-whole relation defining a lower concept as a thing included within a whole thing defined by an upper concept;

means for storing a plurality of conceptual concepts expressing semantic information in said fact information; and

means for inputting other fact information modifying a conceptual concept, and adding exclusion information with at least one subconcept to said at least one conceptual concept so that the expression of said conceptual concept to which said exclusion information with said at least one subconcept has been added, becomes effective to all subconcepts of said conceptual concept except said at least one subconcept.

16. An information storage and retrieval system according to claim 15, wherein said system further comprises:

means for storing said other fact information modifying conceptual concepts and adding said other fact information to said conceptual concepts expressing semantic information in said fact information.

17. An information storage and retrieval system according to claim 15, wherein said system further comprises:

means for storing said exclusion information added to at least one conceptual concept of said conceptual concepts expressing semantic information in said fact information so that said exclusion information expresses exclusion of said at least one conceptual concept from subconcepts of said conceptual concept.

18. An information storage and retrieval system according to claim 14, wherein said system further comprises means for storing said general information added to said conceptual concepts expressing semantic information in said fact information.

19. An information storage and retrieval system according to claim 14, wherein said system further comprises means for storing said indefinite information added to said conceptual concepts for expressing that semantic information in said fact information, is indefinite so that said indefinite information expresses that the expression of said conceptual concepts as semantic information is indefinite.

20. An information storage and retrieval system according to claim 14, wherein said system further comprises means for storing semantic information contained in said fact information, by extracting fact information from other fact information.

21. An information storage and retrieval system comprising:

a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept, a numerical value conceptual concept and a unit conceptual concept, relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

means for inputting fact information represented in terms of said conceptual concepts said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept defining a relation between, and defining a relation between a relational concept or an attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form, wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

a knowledge base for storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

means for generating a retrieval request for a conceptual concept based on said templates, said conceptual concepts and said stored fact information as a result of interaction with a user;

means for retrieving a requested conceptual concept based on said retrieval request;

means for displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form for viewing by said user;

means for storing standard cases to define semantic information in said templates, and an order of semantic priority among standard cases necessary for representing said relational concepts;

means for selecting a conceptual concept used in said fact information and a retrieval condition, extracting candidate sentences according to said templates that use said selected conceptual concept by using said templates, and displaying said candidate sentences;

means for selecting at least one candidate sentence from said candidate sentences;

means for selecting a case from cases in said selected candidate sentences, said case corresponding to said standard cases containing said retrieval condition using said templates from cases in said selected candidate sentence; and

means for substituting said conceptual concept for another conceptual concept in said selected case of said template and displaying said conceptual concept substituted in said selected case.

22. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for inputting semantic information based on a natural language sentence having a plurality of cases selected by a user using a pointing device.

23. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for selecting a conceptual concept used in said fact information and said case displayed in said node tree window by using a pointing device.

24. An information storage and retrieval system according to claim 23, wherein said system further comprises:

means for obtaining an uppermost conceptual concept allowed to enter a case defined in said template by selecting a conceptual concept in said fact information and said retrieval condition by a user using said pointing device, rewriting said conceptual concepts in a tree form in said node tree window to said uppermost conceptual concept and subconcepts in a tree form so that said uppermost conceptual concept stands in the center of said tree form, and substituting a selected conceptual concept for the conceptual concept in said case by selecting said conceptual concept allowed to enter said case by said user using said pointing device.

25. An information storage and retrieval system according to claim 24, wherein said system further comprises:

means for displaying conceptual concepts in the form of a tree so that it is easy to discriminate between conceptual concepts allowed to be replaced and conceptual concepts forbidden to be replaced.

26. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for displaying candidate sentences selected so that said candidate sentences are allowed to be used in terms of the selected conceptual concept or relational concept.

27. An information storage and retrieval system according to claim 26, wherein said system further comprises:

means for displaying said selected candidate sentences so that it is easy to select a case allowed to use the conceptual concept used for the selection of said candidate sentences when selecting said case from cases in said selected candidate sentences.

28. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for generating a retrieval condition for modifying a conceptual concept by selecting said conceptual concept in said retrieval condition and by inputting a conditional sentence, and for displaying said retrieval condition.

29. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for deleting a conditional sentence by selecting said conditional sentence through a pointing device and re-displaying the retrieval condition.

30. An information storage and retrieval system according to claim 21, wherein said system further comprises:

means for extracting cases defined in a template to be used by a conditional sentence in said retrieval condition by selecting said conditional sentence through a pointing device, and displaying cases allowed to be used in said conditional sentence; and

means for performing addition and deletion of cases in said conditional sentence by selecting the displayed case through said pointing device, and displaying the cases.

31. A display method in an information storage and retrieval system comprising the steps of:

providing a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept, a numerical value conceptual concept and a unit conceptual concept, relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

inputting fact information represented in terms of said conceptual concepts said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept, and defining a relation between said relational concept or said attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form, wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

generating a retrieval request for a conceptual concept based on said templates, said conceptual concepts and said stored fact information as a result of interaction with a user;

retrieving a requested conceptual concept based on said retrieval request; and

displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form for viewing by said user.

32. A display method in an information storage and retrieval system comprising the steps of:

providing a concept relation model for representing knowledge in a hierarchical tree form in terms of conceptual concepts including at least a noun conceptual concept, a name conceptual concept, a numerical value conceptual concept and a unit conceptual concept relational concepts each defining relations between plural conceptual concepts and attribute name concepts each defining relation resulting from an attribute common to plural;

wherein said conceptual concepts, relational concepts and attribute name concepts are formed into plural templates including at least a first template which includes two conceptual concepts related to each other by an attribute name concept, and a second template which includes plural conceptual concepts, related to each other by a relational concept;

inputting fact information represented in terms of said conceptual concepts, said relational concepts and attribute name concepts according to said first and second templates defining previously inputted fact information and fact information currently being inputted, each of said templates being related to a relational concept or an attribute name concept, and defining a relation between said relational concept or said attribute name concept and semantic information which includes at least one conceptual concept, wherein said templates are arranged in said hierarchical tree form, wherein a relation between a relational concept or an attribute name concept in an upper template and a relational concept or an attribute name concept in a lower template is defined as an is-a relation and wherein a relation between an upper template and a lower template is defined as an is-a relation;

storing said fact information, said conceptual concepts, said relational concepts, said attribute name concepts and said templates;

generating a retrieval request for a conceptual concept based on said templates, said conceptual concepts and said stored fact information as a result of interaction with a user;

retrieving a requested conceptual concept based on said retrieval request;

displaying a fact window including said inputted fact information and a node tree window including said conceptual concepts, said relational concepts and said attribute name concepts in said hierarchical tree form for viewing by said user;

storing standard cases to define semantic information in said templates, and an order of semantic priority among standard cases necessary for representing said relational concepts;

selecting a conceptual concept used in said fact information and a retrieval condition, extracting candidate sentences according to said templates and allowed to use said selected conceptual concept by using said templates, and displaying said candidate sentences;

selecting at least one candidate sentence from said candidate sentences;

selecting a case from cases in said selected candidate sentences, said case corresponding to said standard cases containing said semantic information of said template and used in said fact information and said retrieval condition using said template; and

substituting said conceptual concept for another conceptual concept in said selected case of said template and displaying said conceptual concept substituted in said selected case.

33. A display method in an information storage and retrieval system according to claim 32, wherein semantic information based on a natural language sentence having a plurality of cases is selected by said user using a pointing device.

34. A display method in an information storage and retrieval system according to claim 32, wherein a conceptual concept used in said fact information and said case of said retrieval condition is selected from conceptual concepts displayed in said node tree window by said user using a pointing device.

35. A display method in an information storage and retrieval system according to claim 34, wherein an uppermost conceptual concept allowed to enter a case defined in said template by selecting the conceptual concept in said fact information and said retrieval condition by a user using said pointing device, rewriting said conceptual concepts in tree form in said node tree window to said uppermost conceptual concept and subconcepts in said tree form so that said uppermost conceptual concept stands in the center of said tree, and substituting a selected conceptual concept for the conceptual concept in said case by selecting said conceptual concept allowed to enter said case by said user using said device.

36. A display method in an information storage and retrieval system according to claim 35, wherein conceptual concepts in a tree form are displayed so that it is easy to discriminate between conceptual concepts allowed to be replaced and conceptual concepts forbidden to be replaced.

37. A display method in an information storage and retrieval system according to claim 32, wherein candidate sentences allowed to be used in terms of the pointed to conceptual concept or relational concept are displayed by focusing (filtering) said candidate sentences.

38. A display method in an information storage and retrieval system according to claim 37, wherein said focused candidate sentences are displayed so that it is easy to select a case allowed to use the conceptual concept used for the selection of said candidate sentences when selecting said case from cases in said selected candidate sentences.

39. A display method in an information storage and retrieval system according to claim 32, wherein a retrieval condition for modifying a conceptual concept is generated by selecting said conceptual concept in said retrieval condition and by inputting a conditional sentence and displaying said retrieval condition.

40. A display method in an information storage and retrieval system according to claim 32, wherein a conditional sentence is deleted by selecting said conditional sentence through a pointing device and the retrieval condition is re-displaying the retrieval condition.

41. A display method in an information storage and retrieval system according to claim 32, wherein cases defined in a template to be used by a conditional sentence in said retrieval condition by selecting said conditional sentence through a user using a pointing device to display cases allowed to be used in said conditional sentence, and wherein additional and deletion of cases in said conditional sentence are performed by selecting the displayed case through said pointing device to display the cases.