System, method, program product, and networking use for recognizing words and their parts of speech in one or more natural languages

  • US 7,680,649 B2
  • Filed: 06/17/2002
  • Issued: 03/16/2010
  • Est. Priority Date: 06/17/2002
  • Status: Active Grant
  • ×
    • Pin Icon | RPX Insight
    • Pin
First Claim
Patent Images

1. A computer-implemented system for recognizing one or more words not listed in a dictionary database, the system comprising:

  • at least one central processing unit;

    a memory operably associated with the at least one processing unit; and

    a dictionary augmentation system storable in memory and executable by the at least one processing unit, the dictionary augmentation system comprising;

    a root process that searches the dictionary database to obtain root information about a root word, the root word being a word with no prefix and suffix; and

    a statistical process that, if the root word is not found in the dictionary database, checks one or more proper substrings of the root word comprising two or more characters in the root word and every proper substring having fewer characters than the root word, against a complete database of each and every possible subset of individual valid words within the dictionary database, to determine, from the likelihood that the proper substring of the root word occurs in a sequence in the subsets of the individual valid words, a probability that the root word is a valid word that was previously unknown, wherein each character in the root word and in the individual valid words is an alphabet-based character and wherein the dictionary database is distinct from the complete database.

View all claims
    ×
    ×

    Thank you for your feedback

    ×
    ×