Work Files Saved Searches
   My Account                                                  Search:   Quick/Number   Boolean   Advanced       Help   


 The Delphion Integrated View

  Buy Now:   Buy PDF- 10pp  PDF  |   File History  |   Other choices   
  Tools:  Citation Link  |  Add to Work File:    
  View:  Expand Details   |  INPADOC   |  Jump to: 
 
 Email this to a friend  Email this to a friend 
       
Title: US5991710: Statistical translation system with features based on phrases or groups of words
[ Derwent Title ]


Country: US United States of America

View Images High
Resolution

 Low
 Resolution

 
10 pages

 
Inventor: Papineni, Kishore Ananda; Ossining, NY
Roukos, Salim Estephan; Scarsdale, NY
Ward, Robert Todd; Croton-on-Hudson, NY

Assignee: International Business Machines Corporation, Armonk, NY
other patents from INTERNATIONAL BUSINESS MACHINES CORPORATION (280070) (approx. 44,393)
 News, Profiles, Stocks and More about this company

Published / Filed: 1999-11-23 / 1997-05-20

Application Number: US1997000859586

IPC Code: Advanced: G06F 17/28;
Core: more...
IPC-7: G06F 17/28;

ECLA Code: G06F17/28D2;

U.S. Class: Current: 704/002; 704/009;
Original: 704/002; 704/009;

Field of Search: 704/002-7,9

Government Interest:     The U.S. Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others on reasonable terms as provided for by the terms of Contract No. N00014-92-C-0189 awarded by the Office of Naval Research.

Priority Number:
1997-05-20  US1997000859586

Abstract:     A system for translating a first word set in a source language into a second word set in a target language, the system comprising: input means for inputting the first word set into the system; tagging means for tagging the first word set input to the system so as to at least substantially reduce non-essential variability in the first word set; translation means including a single a posteriori conditional probability model and a target candidate store for storing target language candidate word sets, wherein the translation means employs the single model to evaluate the target language candidate word sets in order to select the target language candidate word set having a best score with respect to the first word set; and output means for outputting the best scoring target language candidate word set as the second word set in the target language.

Attorney, Agent or Firm: F. Chau & Associates, LLP ;

Primary / Asst. Examiners: Isen, Forester W.; Edouard, Patrick N.

INPADOC Legal Status: Show legal status actions

Family: None

First Claim:
Show all 27 claims
What is claimed is:     1. A system for translating a first word set in a source language into a second word set in a target language, the system comprising:
  • input means for inputting the first word set into the system;
  • tagging means for tagging the first word set input to the system so as to at least substantially reduce non-essential variability in the first word set;
  • translation means including a single a posteriori conditional probability model and a target candidate store for storing target language candidate word sets, wherein the translation means employs the single model to evaluate the target language candidate word sets in order to select the target language candidate word set having a best score with respect to the first word set;
  • wherein the single model includes a prior model, a plurality of feature functions and a plurality of weighting factors respectively corresponding to the plurality of feature functions, wherein the translation means receives the first word set and, in accordance with the single model, iteratively proceeds through the target language candidate store and each of the candidate word sets and finds the feature functions which are true with respect to the first word set and the second word set, and then multiplies the prior model by the weighting factors corresponding to the feature functions thereby providing resulting scores, the translation means then choosing the best matching target language candidate word set in accordance with the resulting scores; and
  • output means for out putting the best scoring target language candidate word set as the second word set in the target language.


Background / Summary: Show background / summary

Drawing Descriptions: Show drawing descriptions

Description: Show description

Forward References: Show 130 U.S. patent(s) that reference this one

       
U.S. References: Go to Result Set: All U.S. references   |  Forward references (130)   |   Backward references (7)   |   Citation Link

Buy
PDF
Patent  Pub.Date  Inventor Assignee   Title
Get PDF - 22pp US4773039  1988-09 Zamora  International Business Machines Corporation Information processing system for compaction and replacement of phrases
Get PDF - 29pp US5237502  1993-08 White et al.  International Business Machines Corporation Method and apparatus for paraphrasing information contained in logical forms
Get PDF - 15pp US5293584  1994-03 Brown et al.  International Business Machines Corporation Speech recognition system for natural language translation
Get PDF - 16pp US5424947  1995-06 Nagao et al.  International Business Machines Corporation Natural language analyzing apparatus and method, and construction of a knowledge base for natural language analysis
Get PDF - 111pp US5477451  1995-12 Brown et al.  International Business Machines Corp. Method and system for natural language translation
Get PDF - 75pp US5510981  1996-04 Berger et al.  International Business Machines Corporation Language translation apparatus and method using context-based translation models
Get PDF - 13pp US5568383  1996-10 Johnson et al.  International Business Machines Corporation Natural language translation system and document transmission network with translation loss information and restrictions
       
Foreign References: None

Other References:
  • A Maximum Entropy Approach to Natural Language Processing, Berger et al., Association for Computation Linguistics, pp. 39-71, vol. 22, No. 1, Mar. 1996. (33 pages) Cited by 8 patents [ISI abstract]
  • A Maximum Entropy Model for Parsing, Ratnaparkhi et al., Proceedings of International Conference on Spoken Language Processing, Yokohama, Japan, pp. 803-806, vol. 2, 1994.
  • Decision Tree Models Applied to the Labeling of Text with Parts-of-Speech, Black et al., Proceedings of Speech and Natural Language Workshop, pp. 117-121, New York, Feb. 1992.
  • Decision Tree Parsing Using a Hidden Derivative Model, Jelinek et al., Proceedings of Human Language Technologies Workshop, Plainsboro, N.J., pp. 272-277, Mar. 8-11, 1994.
  • Part of Speech Assignment By a Statistical Decision Algorithm, Bahl et al., IEEE International Symposium on Information Theory, pp. 88-89, Ronneby, Sweden (1976).
  • Recent Progress in Hidden Understanding Models, Miller et al., Proceedings of Spoken Language Systems Workshop, pp. 276-280, Austin, Jan. 1995.
  • Statistical Source Channel Models for Natural Language Understanding, Mark E. Epstein, Ph.D. Thesis, New York University, Sep. 1996.
  • The Application of Semantic Classification Trees to Natural Language Understanding, Kuhn t al., IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 449-460, vol. 17, No. 5, May 1995. (12 pages) Cited by 7 patents [ISI abstract]
  • The Mathematics of Statistical Machine Translation: Parameter Estimation, Brown et al., Association for Computational Linguistics, pp. 263-311, vol. 19, No. 2, Jun. 1993.
  • Chronus, The Next Generation, Levin et al., Speech Research Department, AT&T Bell Laboratories, pp. 269-271.


  • Inquire Regarding Licensing

    Powered by Verity


    Plaques from Patent Awards      Gallery of Obscure PatentsNominate this for the Gallery...

    Thomson Reuters Copyright © 1997-2014 Thomson Reuters 
    Subscriptions  |  Web Seminars  |  Privacy  |  Terms & Conditions  |  Site Map  |  Contact Us  |  Help