Apparatus and method for presenting document data

• US 6,912,536 B1
• Filed: 07/23/1999
• Issued: 06/28/2005
• Est. Priority Date: 12/04/1998
• Status: Expired due to Fees

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Ochitani, Ryo
• Primary Examiner(s): Ali, Mohammad
• Application Number:
  US09/360,487
• Time in Patent Office: 2,167 Days
• Field of Search: 707 1- 10, 707200-205, 707100-1541, 709/9, 709/219, 345/810, 345/700, 715/514
• US Class Current: 707/737
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: Y10S 707/917 Text
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A word id extracted from dated document data, and the number of occurrences of the word can be obtained by adding up the number of the extracted word for each field and period. Furthermore, a word indicating a large number of occurrences in each field and period is extracted as a characteristic word. When a user specifies a field and a period, characteristic words in the document data in the specified period are displayed. When the user selects a specific characteristic word, the header, etc. of the document data containing the characteristic word is displayed.

Claims

1. A document data presentation apparatus, comprising:

a document feature information extraction unit calculating a number of occurrences of each of some words in document data and extracting and synthesizing document feature information composed of words and the number of occurrences of each word;

a classification unit classifying document data into respective fields by comparing a preliminarily calculated field property vector of each of a plurality of fields and a text property vector corresponding to the document feature information; and

a document information presentation unit extracting document data whose generation date is in a predetermined period in a same field or which contains a date in the predetermined period in a same field and displaying a number of extracted document data as a time series graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when a user designates an arbitrary period of a graph.

2. The apparatus according to claim 1, further comprising:

a field feature information generation unit setting document feature information composed of some words contained in reference document data whose field is specified and the number of occurrences of each word as the field feature information of a corresponding field; and

a field classification unit calculating similarity between document feature information of document data to be classified and field feature information generated by the field feature information generation unit and classifying document data whose similarity exceeds a predetermined value into a field designated by the field feature information.

3. The apparatus according to claim 1, further comprising:

a field feature information generation unit extracting document feature information composed of some words contained in referenced document data having a user feature similar to a feature of a user that retrieves document data and the number of occurrences of each word and setting the document feature information as its field feature information; and

a field classification unit calculating similarity between document feature information of document data to be classified and field feature information generated by the field feature information generation unit and classifying document data whose similarity exceeds a predetermined value into the field.

4. A document data presentation apparatus, comprising:

a document feature information extraction unit calculating a number of occurrences of some words in document data and extracting and synthesizing document feature information composed of the words and the number of occurrences of each word;

a field classification unit calculating similarity between a text property vector corresponding to document feature information extracted from the document feature information extraction unit and a preliminarily calculated field property vector corresponding to document data whose field is specified and classifying document data whose similarity exceeds a predetermined value into the field; and

a document information presentation unit extracting document data whose generation is in a predetermined period or which contains a date in the predetermined period in a same field and displaying a number of extracted document data as a time series trend graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when a user designates an arbitrary period of the graph.

5. The apparatus according to claim 4, further comprising:

a field feature information generation unit handling document feature information generation unit handling document feature information as a cluster, calculating similarity among clusters, uniting clusters whose similarity is high and setting the document feature information of each cluster as field feature information when a number of clusters falls below a predetermined value, and

a field classification unit calculating similarity between document feature information of document data to be classified and field feature information generated by the field feature information generation unit and classifying document data whose similarity exceeds a predetermined value into the field.

6. The document data presentation apparatus according to claim 4, wherein

said field classification unit means generates a partial document set comprising a set of document data in each field and period based on a result of cluster classification obtained in an automatic clustering process, compares a time series change of statistic information about the partial document set with a time series change of statistic information about a set of document data in each predetermined field and period, and associates a field classified by the automatic clustering process with the predetermined field.

7. The document data presentation apparatus according to claim 4, wherein

if an arbitrary word is selected as the document feature information when words whose number of occurrences exceeds a predetermined value are displayed, said document information presentation unit displays a list of document data containing the selected word.

8. The data presentation apparatus according to claim 4, wherein

said field classification unit further comprises a field estimation unit estimating a field related to the document data to be classified by calculating similarity between the time series change of statistical information about the document feature information of a set of document data in each predetermined period of a known field and the time series change of statistical information about the document feature information to be classified.

9. The document data presentation apparatus according to claim 4, wherein

said document feature information extraction unit means obtains statistic information about a weighted word from document data in each field and period, and extracts a characteristic word as the feature information according to the statistic information.

10. The data presentation apparatus according to claim 4, wherein

said document feature information extraction unit means obtains a number of occurrences of a word in the set of document data in each period as statistic information about the characteristic word, and extracts a word indicating a large number of occurrences as the characteristic word.

11. The document data presentation apparatus according to claim 4 wherein

said document information presentation unit statistic information obtaining means obtains a number of pieces of document data in a same field in each period or a ratio of a number of pieces of document data in each period to a number of pieces of all document data in each period as statistic information about the set of the document data.

12. The document data presentation apparatus according to claim 4, further comprising:

a user document designation history information storage unit storing user feature information of a user as associated with document identification information about document data referred to by the user; and

a classification unit classifying the document data according to the user feature information of the user that has referred to the document data.

13. The apparatus according to claim 12, further comprising

access information obtaining means for obtaining access information about document data referred to by a specific user, and access information about document data referred to by a plurality of users sharing same user features with the specific user.

14. A document data presentation apparatus, comprising:

a user document designation history information storage unit storing user feature information of a user as associated with document identification information indicating document data referenced by the user;

a classification unit classifying the document data into respective fields by comparing a preliminarily calculated field property vector for each of a plurality of fields and a text property vector corresponding to the user feature information of the user that has referred to the document data;

a document feature information extraction unit extracting and synthesizing document feature information composed of some words and a number of occurrences of each word from among a set of document data whose generation date is in a predetermined period in a same field of which contains a date in the predetermined period in a same field; and

a document information presentation unit displaying a number of document data in a predetermined period as a time series trend graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when a user designates an arbitrary period of the graph.

15. A document data presentation system in which a document data presentation apparatus is connected to a display device for displaying document data through communications circuit, wherein:

said document data presentation apparatus comprises;

a document feature information extraction unit calculating a number of occurrences of each of some words in document data and extracting and synthesizing document feature information composes of the words and the number of occurrences of each word;

a classification unit classifying document data into respective fields by comparing a preliminarily calculated field property vector of each of a plurality of fields and a text property vector corresponding to the document feature data; and

a transmission unit extracting document data whose generation date is in a predetermined period in a same field or which contains a date in the predetermined period in a same field and displaying extracted document data as a time series trend graph, and simultaneously transmitting document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when a user designates an arbitrary period of the graph; and

said display device comprises

a reception unit receiving information from the said document data presentation apparatus; and

a display control unit displaying a number of extracted document data in a predetermined period in a same field as a time series trend graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when an arbitrary period of the graph is designated.

18. The method according to claim 15, further comprising the steps of:

generating a first partial document set by classifying document data into respective fields according to field information preliminarily set by a user, and generating a second partial document set by classifying document data into respective fields by an automatic clustering process; and

comparing a time series change of statistic information between the first and second partial document sets, and associating the fields classified by the automatic clustering process with the fields classified according to field information set by the user.

16. A document data presenting method comprising:

calculating a number of occurrences of each of some words in document data and extracting and synthesizing document feature information composed of the words and the number of occurrences of each word;

classifying document data into respective fields by comparing a preliminarily calculated field property vector of each of a plurality of fields and a text property vector corresponding to the document feature information; and

extracting document data whose generation date is in a predetermined period in a same field or which contains a date in the predetermined period in a same field and displaying a number of extracted document data as a time series trend graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period when a user designates an arbitrary period of the graph.

17. The document data presenting method according to claim 16, wherein

said document feature information extraction unit extracts words whose number of occurrences exceeds a predetermined value as characteristic words.

19. Document data presenting method according to claim 16, further comprising:

storing user feature information of a user as associated with document identification information about document data referred to by the user; and

classifying the document data according to the user feature information of the user who has referred to the document data.

20. The document data presenting method according to claim 16, further comprising:

storing user feature information of a user as associated with document identification information indicating document data referenced by the user; and

classifying the document data according to the user feature information of the user who has referred to the document data.

21. Computer-readable storage medium storing a program for performing a process, said process comprising:

calculating a number of occurrences of each of some words in document data and extracting and synthesizing document feature information composed of the words and the number of occurrences of each word;

classifying document data into respective fields by comparing a preliminarily calculated field property vector of each of a plurality of fields and a text property vector corresponding to the document feature information; and

extracting document data whose generation date is in a predetermined period in a same field or which contains a date in the predetermined period in a same field and displaying a number of extracted document data as a time series trend graph, and simultaneously displaying document feature information whose number of occurrences of each word in a designated period exceeds a predetermined value when a user designates an arbitrary period of the graph.

22. The computer-readable storage medium according to claim 21, said process further comprising

storing user feature information of a user as associated with document identification information about document data referred to by the user;

and

classifying the document data according to the user feature information of the user who has referred to the document data.

23. The computer-readable storage medium according to claim 21, said process further comprising:

storing user feature information of a user as associated with document identification information indicating document data referenced about document data referred to by the user;

classifying the document data according to the user feature information of the user who has referred to the document data.