Text management system

• US 5,940,624 A
• Filed: 09/26/1994
• Issued: 08/17/1999
• Est. Priority Date: 02/01/1991
• Status: Expired due to Term

• Current Assignee: Miles, Inc. (Bayer AG), Intel Corporation, Wang Laboratories Incorporated
• Original Assignee: Wang Laboratories Incorporated
• Inventors: Kadashevich, A. Julie, Harvey, Mary F., Clark, Cheryl
• Primary Examiner(s): Hofsass, Jeffery
• Assistant Examiner(s): HILL, ANDREW M
• Application Number:
  US08/094,033
• Time in Patent Office: 1,786 Days
• Field of Search: 395/751, 395/758, 395/759, 395/760, 395/793, 395/794, 395/795, 395/796, 395/798
• US Class Current: 704/9
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates

Abstract

An first apparatus for searching a collection of words based upon an input word, the first apparatus including means for generating a first set of words containing members that are lexically related to the input word, the first set of words including words that are other than regular inflectional nouns; and a search engine for searching the collection of words to detect the occurrence of any of the words from a group of search words, the group of search words comprising the input word and the first set of words. Also a second apparatus for generating a plurality of topic words from the collection of words, that apparatus including means for selecting a subset of words from the collection of words; a morphological analyzer for generating morphological information about each of the words in the subset of words; means for evaluating whether a given word of the subset of words conveys information about the content of the collection of words, the evaluation means basing its determination upon the morphological information generated by the morphological analyzer; and means for generating a topic word corresponding to the given word, if the evaluation means determines that the given word conveys information about the content of the collection of words.

Claims

1. An apparatus for generating a plurality of words based upon and lexically related to an input word, comprising:

a morphological analyzer for parsing the input word to determine at least one stem of the input word having a meaning lexically related to the meaning of the input word and, for each stem of the input word, one or more suffixes sequentially attached to the stem; and

a word generation means responsive to the morphological analyzer for generating a first group of words from the input word that have meanings lexically related to the meaning of the input word by stripping successive suffixes from the input word,

wherein each word of the first group of words is comprised of the input word with one or more suffixes removed, and

wherein the suffixes stripped from the input word are selected from a first set of suffixes.

2. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 1, wherein:

the suffixes of the first set of suffixes include only suffixes which do not substantially change the meaning of words when stripped from words.

3. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 2, wherein the first set of suffixes include the suffixes:

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly, #ize, +ify, +fy and #y.

4. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 1, wherein:

the first set of words includes both words that are regular inflectional nouns and words that are other than regular inflectional nouns.

5. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 1, wherein the word generation means further comprises:

a means for generating additional words of the first group of words by adding suffixes selected from a second group of suffixes to the words of the first group of words wherein

the suffixes of the second group of suffixes include only suffixes which do not substantially change the meaning of words when added to words.

6. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 5, wherein the second group of suffixes comprise:

both inflectional and derivational suffixes, including the suffixes

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly and #y.

7. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 1, further comprising:

a thesaurus database; and

in the word generation means, a means for indexing the thesaurus database with words of the first group of words to generate from the thesaurus database a second group of words, wherein

the words of the second group of words are synonyms of words of the first set of words.

8. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 7, wherein the word generation means further comprises:

a means for generating additional words of the second group of words by adding suffixes selected from a second group of suffixes to the words of the second group of words wherein

the suffixes of the second group of suffixes include only suffixes which do not substantially change the meaning of words when added to words.

9. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 8, wherein the second group of suffixes include the suffixes:

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly and #y.

10. The apparatus for generating a plurality of words based upon and lexically related to an input word of claim 1, wherein said morphological analyzer further comprises:

a recognition engine for generating a derivational history for each parsing of the input word into a stem and one or more suffixes, wherein

each derivational history includes the part of speech of the word generated by the corresponding parse of the input word.

11. A method for generating a plurality of words based upon and lexically related to an input word, comprising the steps of:

parsing the input word to determine at least one stem of the input word having a meaning lexically related to the meaning of the input word and, for each stem of the input word, one or more suffixes sequentially attached to the stem; and

generating a first group of words from the input word that have meanings lexically related to the meaning of the input word by stripping successive suffixes from the input word,

wherein each word of the first group of words is comprised of the input word with one or more suffixes removed, and

wherein the suffixes stripped from the input word are selected from a first set of suffixes.

12. The method for generating a plurality of words based upon and lexically related to an input word of claim 11, wherein:

the suffixes of the first set of suffixes include only suffixes which do not substantially change the meaning of words when stripped from words.

13. The method for generating a plurality of words based upon and lexically related to an input word of claim 12, wherein the first set of suffixes include the suffixes:

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly, #ize, +ify, +fy and #y.

14. The method for generating a plurality of words based upon and lexically related to an input word of claim 11, wherein:

the first set of words includes both words that are regular inflectional nouns and words that are other than regular inflectional nouns.

15. The method for generating a plurality of words based upon and lexically related to an input word of claim 11, further comprising the steps of:

generating additional words of the first group of words by adding suffixes selected from a second group of suffixes to the words of the first group of words wherein

the suffixes of the second group of suffixes include only suffixes which do not substantially change the meaning of words when added to words.

16. The method for generating a plurality of words based upon and lexically related to an input word of claim 15, wherein the second group of suffixes include the suffixes:

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly and #y.

17. The method for generating a plurality of words based upon and lexically related to an input word of claim 11, further comprising the steps of:

indexing a thesaurus database with words of the first group of words to generate from the thesaurus database a second group of words, wherein

the words of the second group of words are synonyms of words of the first set of words.

18. The method for generating a plurality of words based upon and lexically related to an input word of claim 17, further comprising the steps of:

generating additional words of the second group of words by adding suffixes selected from a second group of suffixes to the words of the second group of words

wherein

the suffixes of the second group of suffixes include only suffixes which do not substantially change the meaning of words when added to words.

19. The method for generating a plurality of words based upon and lexically related to an input word of claim 18, wherein the second group of suffixes include the suffixes:

#ful, #ish, +ous, +ic, +al, #ar, #er, #or, +ive, +ory, #able, +able, +ible, #ment, #ness, +ity, +ety, +ty, #ly and #y.

20. The method for generating a plurality of words based upon and lexically related to an input word of claim 11, further comprising the step of:

generating a derivational history for each parsing of the input word into a stem and one or more suffixes, wherein

each derivational history includes the part of speech of the word generated by the corresponding parse of the input word.