Document processing system and recording medium

• US 6,523,025 B1
• Filed: 08/01/2000
• Issued: 02/18/2003
• Est. Priority Date: 03/10/1998
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Hashimoto, Minako, Kashino, Wakako, Ochitani, Ryo, Nishino, Fumihito
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): Rayyan, Susan F.
• Application Number:
  US09/630,553
• Time in Patent Office: 931 Days
• Field of Search: 707/1, 707/3, 707/5
• US Class Current: 1/1
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

The accuracy of retrieving or clipping documents is improved. A document to be processed is input via a document input section. Event specifying means looks up knowledge information stored in knowledge information storing means to specify the type of an event described in the input document. Attribute value extracting means extracts, from the document, attribute values of attributes relating to the specified event. Correlating means performs a process of correlating the attribute values extracted by the attribute value extracting means with entities in the real world. Document storing means stores information (normalized information) generated by the correlating means and the document or information specifying a storage location thereof in a manner associated with each other. Document extracting means compares a query input from a user interface section with the normalized information and extracts, from the document storing means, matching documents or information specifying their storage locations.

Claims

1. A document processing system for storing input documents after subjecting the documents to a predetermined process, and for retrieving or clipping documents matching a given query from the stored documents, comprising:

knowledge information storing means for storing knowledge information used while processing an input document, where the processing comprises specifying, extracting, and correlating contents of the input document, and where the knowledge information comprises patterns describing contexts of text within which types of events are expected to be found;

event specifying means for specifying a type of an event described in the input document by looking up patterns in said knowledge information storing means and by matching text in the input document to a pattern that corresponds to the specified event type;

attribute value extracting means for extracting, from the document, attribute values of attributes relating to the event specified by said event specifying means by looking up the knowledge information stored in said knowledge information storing means;

correlating means for correlating the attribute values extracted by said attribute value extracting means with attributes associated with entities in the real world that are stored in said knowledge information storing means;

document storing means for storing the attribute values correlated by said correlating means and the document or information specifying a storage location thereof in a manner associated with each other; and

document extracting means for retrieving or clipping a target document by looking up the attribute values and the query.

2. The document processing system according to claim 1, wherein, if the extracted attribute values include a proper name, said correlating means specifies a unique entity in the real world that is represented by the proper name by matching other of the attribute values to a subset of the stored attributes that are associated with the specified entity, and assigns to the proper name predetermined information uniquely identifying the specified entity, and

said document extracting means looks up the predetermined information assigned by said correlating means to perform retrieval or clipping.

3. The document processing system according to claim 1, wherein, if the attribute values include a reference expression, said correlating means specifies an attribute value which the reference expression refers to.

4. The document processing system according to claim 3, further comprising importance calculating means for calculating a degree of importance of a target document by looking up a frequency of occurrence of a keyword included in the document,

said importance calculating means equally processing the keyword and the reference expression whose target of reference is specified by said correlating means.

5. The document processing system according to claim 1, further comprising normalizing means for converting a numerically convertible attribute value, among the attribute values, to a corresponding numerical value, thereby normalizing the numerically convertible attribute value, and wherein

said document extracting means looks up information normalized by said normalizing means to perform retrieval or clipping.

6. The document processing system according to claim 5, further comprising unit converting means for converting a unit of the numerical value obtained by said normalizing means into a predetermined unit.

7. The document processing system according to claim 5, further comprising tense acquiring means for acquiring tense of a predetermined sentence constituting the document, and

creation date acquiring means for acquiring a date of creation of the document, and wherein

said normalizing means looks up the tense of the document acquired by said tense acquiring means and the creation date acquired by said creation date acquiring means, to estimate a definite value of an attribute value indicating a date or a term.

8. The document processing system according to claim 7, further comprising importance calculating means for calculating a degree of importance of a target document by looking up a frequency of occurrence of a keyword included in the document,

said importance calculating means calculating the degree of importance taking account of the date or term estimated by said normalizing means.

9. The document processing system according to claim 1, wherein said event specifying means, said attribute value extracting means and said correlating means process the query in a manner similar to that in which the document is processed, and

said document extracting means looks up the attribute values of the document and of the query correlated by said correlating means, to perform retrieval or clipping.

11. The apparatus according to claim 1, wherein the knowledge information used for processing input documents describes patterns of content that are expected to be found in a subject domain of the input documents.

10. A computer-readable recording medium recording a program for causing a computer to perform a process of storing input documents after subjecting the documents to a predetermined process and retrieving or clipping documents matching a given query from the stored documents, wherein the program causes the computer to function as

knowledge information storing means for storing knowledge information necessary for processing an input document, and where the knowledge information comprises patterns describing contexts of text within which types of events are expected to be found;

event specifying means for specifying a type of an event described in the input document by looking up patterns stored in the knowledge information storing means,

attribute value extracting means for extracting, from the document, attribute values of attributes relating to the event specified by the event specifying means by looking up the knowledge information stored in the knowledge information storing means, and by matching text in the input document to a pattern that corresponds to the specified event type;

correlating means for correlating the attribute values extracted by the attribute value extracting means with attributes associated with entities in real world that are stored in the knowledge information storing means,

document storing means for storing the attribute values correlated by the correlating means and the document or information specifying a storage location thereof in a manner associated with each other, and

document extracting means for retrieving or clipping a target document by looking up the attribute values and the query.

12. A method for storing input documents after subjecting the documents to a predetermined process, and for retrieving or clipping documents matching a given query from the stored documents, comprising:

isolating an event and specifying a type of the event by parsing an input document using pre-stored rules comprising expressions that map event types to expressions expected in a subject domain of the input documents;

extracting, from the input document, a value of an attribute associated with the type of the isolated event, by looking up pre-stored knowledge information of the type of the event;

uniquely identifying a single specific real world entity that is named by the extracted value, by looking up pre-stored knowledge information of the entity;

storing the extracted value in association with the entity and the input document; and

retrieving the stored input document by looking up the attribute value and the query.

13. An apparatus for storing input documents after subjecting the documents to a predetermined process, and for retrieving or clipping documents matching a given query from the stored documents, comprising:

an isolating unit isolating an event and specifying a type of the event by parsing an input document using pre-stored rules comprising expressions that map event types to expressions expected in a subject domain of the input documents;

an extracting unit extracting, from the input document, a value of an attribute associated with the type of the isolated event, by looking up pre-stored knowledge information of the type of the event;

an entity identifying unit uniquely identifying a single specific real world entity that is named by the extracted value, by looking up pre-stored knowledge information of the entity;

a storing unit storing the extracted value in association with the entity and content of the input document; and

a retrieving unit retrieving the content of the input document by looking up the attribute value and the query.

14. A document processing system with a program for storing input documents after subjecting the documents to a predetermined process, and with a program for retrieving or clipping documents matching a given query from the stored documents, comprising:

a knowledge information storing unit storing a knowledge base used while processing an input document, where the processing comprises specifying, extracting, and correlating contents of the input document;

an event specifying unit specifying a type of an event described in the input document by looking up the knowledge base;

an attribute value extracting unit extracting, from the document, attribute values of attributes relating to the event specified by said event specifying unit by looking up the knowledge base;

a correlating unit for correlating the attribute values extracted by said attribute value extracting unit with entities in the real world by looking up the knowledge base;

a document storing unit for storing the attribute values correlated by said correlating unit in association with the document or information specifying a storage location thereof; and

a document extracting unit for retrieving or clipping a target document by looking up the attribute values and the query.