System for correcting grammer based parts on speech probability

• US 5,537,317 A
• Filed: 06/01/1994
• Issued: 07/16/1996
• Est. Priority Date: 06/01/1994
• Status: Expired due to Term

• Current Assignee: Binary Services LLC (Intellectual Ventures LLC)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Schabes, Yves, Roche, Emmanuel, Golding, Andrew R.
• Primary Examiner(s): Weinhardt, Robert A.
• Application Number:
  US08/252,106
• Time in Patent Office: 776 Days
• Field of Search: 364/419.01, 364/419.04, 364/419.08, 364/419.12, 364/419.10
• US Class Current: 704/9
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique

Abstract

In a grammar checking system in which a sentence is first tagged as to parts of speech, the probability of the sequence of the parts of speech being correct is utilized to correct improper use of troublesome words, especially those identical sounding words which are spelled differently. The system corrects word usage based not on the probability of the entire sentence being correct but rather on the probability of the sequence of the parts of speech being correct. As part of the subject invention, the parts of speech sequence probability is utilized in part of speech sequence verification, underlying spelling recovery, auxiliary verb correction, determiner correction, and in a context-sensitive dictionary lookup.

Claims

1. In a grammar checking system in which an input sentence is first tagged as to parts of speech, apparatus for correcting word usage in said input sentence, comprising:

means for ascertaining the probability of the part of Speech tag sequence of a candidate word in the input sentence and an easily confused word in a second sentence, said second sentence containing said easily confused word being a candidate sentence;

said means including;

means for detecting that input or candidate sentence having the higher part of speech tag sequence probability to permit establishing which of said two words is correct; and

,

means for selecting as the correct word the word in that input or candidate sentence having the highest part of speech tag sequence probability;

wherein said detecting means includes means for establishing a geometric average of the probability of part of speech tag sequences and wherein said means for selecting selects said correct word based on said geometric average of the probabilities of the parts of speech sequences of said input and candidate sentences.

2. The system of claim 1 wherein said means for ascertaining the probability of said parts of speech sequence includes means for analyzing overlapping tripples of parts of speech.