← Back to Patent Explorer
EP0562334A2 Expired / Lapsed EP ⏱️ 0.0 years remaining

Computer system and method for the automated analysis of texts

Assignee: International Business Machines Corp
•
Filing: 1993-03-05
•
Granted: —
•
Expires: 2013-03-05
📄 Original PDF 🌐 Google Patents 🔄 Sync Now

Abstract

Abstract The invention concerns a computer system for the automated analysis of texts composed of natural language words. The memory of a computer system stores the stems (110) of natural language verbs along with the associated paradigms (111) and endings (112). With the aid of the paradigms and endings, the various forms of the associated verb may be derived from the stem. During the analysis of a text, the stem of a verb occurring therein is looked up in the memory of the computer system. A comparison of the verb to be analyzed with the associated paradigms and endings yields the different verb forms with regard to person and tense. During the analysis of further words of the text, these different forms are considered. As a result, the decision as to which form is the correct one may be delayed for as long as possible. Multiple analysis is eliminated, thus reducing the time required for text analysis.

Inventors

  • 👤 Magdalena Zoeppritz

Bibliographic Details

Application Number —
Jurisdiction EP
Data Source google_patents
Last Synchronized 2026-11-01 17:10
CPC Classifications
G; G06; G06F; G06F40/00; G06F40/20; G06F40/253; G06F40/237; G06F40/247

Claims Analysis

Total Claims: 12
Claim 1
Independent Claim 37 words
Computer system for the automated analysis of texts composed of natural language words, characterized in that the common stem of particular words is stored once, and that for each stem paradigms and endings are stored (Fig. 1).
Claim 2
Dependent Claim 26 words
Computer system as claimed in claim 1, characterized in that the paradigms and the endings are stored in each case in the form of a code.
Claim 3
Dependent Claim 22 words
Computer system as claimed in claim 2, characterized in that tables are stored for associating the codes with the paradigms and endings.
Claim 4
Dependent Claim 27 words
Computer system as claimed in any one of the claims 1 to 3, characterized in that each stem is associated with one of the paradigms (Fig. 2).
Claim 5
Dependent Claim 27 words
Computer system as claimed in any one of the claims 1 to 4, characterized in that each stem is associated with a set of endings (Fig. 3).
Claim 6
Dependent Claim 25 words
Computer system as claimed in any one of the claims 1 to 5, characterized in that the stems associated with the verbs are stored once.
Claim 7
Dependent Claim 28 words
Computer system as claimed in claim 6, characterized in that the paradigms are composed of the stem as applicable to particular tenses and persons of the respective verb.
Claim 8
Dependent Claim 33 words
Computer system as claimed in claim 6 or 7, characterized in that the endings are composed of the endings suffixed to the stem for the various tenses and persons of the respective verb.
Claim 9
Dependent Claim 27 words
Computer system as claimed in any one of the claims 1 to 8, characterized by its use in conjunction with speech recognition, text correction/processing or text translation.
Claim 10
Dependent Claim 40 words
Method for the automated analysis of texts composed of natural language words by means of a computer system, characterized in that the words are compared with a plurality of stored stems and with the respective associated stored paradigms and endings.
Claim 11
Dependent Claim 45 words
Method as claimed in claim 10, characterized in that in response to a single match of a word with one of the stored stems, the asscociated stored paradigms and the associated stored endings, further information concerning the word is derived from these paradigms and endings.
Claim 12
Dependent Claim 43 words
Method as claimed in claim 10, characterized in that in response to several matches of a word with one of the stored stems, the associated paradigms and the associated stored endings, further information concerning the existing word is obtained by analyzing further words.

References Cited (Backward Citations)

No citation records indexed.

USPTO Annuity & Maintenance Windows

Maintenance schedule is calculated for granted US utility patents.

Assigned Categories & Tags

No categories assigned to this patent yet.

Assign Category Tag