Data processing: speech signal processing – linguistics – language – Linguistics – Dictionary building – modification – or prioritization
Reexamination Certificate
1999-03-24
2001-01-09
Thomas, Joseph (Department: 2747)
Data processing: speech signal processing, linguistics, language
Linguistics
Dictionary building, modification, or prioritization
C704S009000, C707S793000, C707S793000
Reexamination Certificate
active
06173253
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an apparatus for allowing a user to input long words in a sentence in terms of elliptic characters without disturbing the continuity of thought. The apparatus according to the present invention is beneficial for increasing the speed and operability of inputting characters by way of a keyboard. It is also applicable for effecting an increase in the input speed when using handwritten character recognition or speech recognition and contributes to the increase in operability of the equipment.
When inputting sentences using a word processor, it is often experienced that words related to private affairs, such as a job and a hobby and to a person's own name are repeatedly input. Especially in a case where those often-used character strings are long, it is a burden for the user to input repeatedly identical, long character strings.
When using an apparatus which allows the user to input words by handwriting with a pen and tablet, since false recognition of characters input by the user may occur, the user has an increased burden in a case in which he or she inputs those characters and long sentences repeatedly.
There is an apparatus that allows the user to input characters or sentences portions omitted partially in order to reduce the user's burden.
For example, in Japanese Patent Application Laid-Open Number 7-191986 (1995) a technology which is disclosed which predicts an intended word and interpolates omitted characters by referring to memories storing syntax coding rules and word usage examples, when the user inputs a sentence including words with omitted characters. On the other hand, in Japanese Patent Application Laid-Open Number 5-28180 (1993) a technology is disclosed which prepares a table storing combinations of adjacent words, such as noun class—verb class and verb class—verbal phrase, and interpolates omitted characters and predicts an intended word by using this table.
As shown in the conventional technologies described above, word-to-word relation information between adjacent words is required to interpolate a sentence including omitted characters. For example, m syntax coding rules and word usage examples are used as this information in Japanese Patent Application Laid-Open Number 7-191986 (1995), and combinations of adjacent words are used as this information in Japanese Patent Application Laid-Open Number 5-28180 (1993).
It is, however, necessary to prepare such word-to-word relation information by referring to a vast amount of reference sentences, and it is not easy to prepare this information only by manual work.
The conventional technologies described above assume that a single word or character in a sentence is omitted, and does not mention the case that a sentence with plural words and/or characters omitted is interpolated.
SUMMARY OF THE INVENTION
An object of the present invention is to provide an apparatus for interpolating a sentence in which plural words and/or characters are omitted.
Another object of the present invention is to provide an apparatus for extracting word-to-word relation information automatically and for preparing a dictionary.
The above object can be attained by a document or sentence processing apparatus having an input unit for inputting characters, a display unit for displaying input characters and a processing unit for converting and editing the input characters, in which the processing unit includes a candidate word extraction means which extracts candidates for the words with their characters omitted and/or omitted words themselves by referring to a vocabulary dictionary storing words and their usage frequency, to a dictionary of the transition between words defining information on the transition between words and the probability of the transition between words, and by searching the characters before and after the elliptic character included in the input sentence into an vocabulary dictionary, and a determination means which selects a single word among the extracted candidate words by referring to the dictionary of transition between words.
The above object can be attained by steps including a step of decomposing the input sentence into single words and storing coordinated pairs of the individual word and its occurrence count, a step of searching the class of the particle for the individual word and storing the count of transition between words into the transition dictionary, a step of extracting candidates for the words with their characters omitted and/or omitted words themselves by focusing on the characters before and after an elliptic character included in the input sentence and searching the vocabulary dictionary, a step of selecting a single word among the extracted candidate words by referring to the dictionary of transition between words, and a step of modifying the occurrence count of the selected word and modifying the transition dictionary on the basis of the information on transition between words in case the selected word is found in the vocabulary dictionary.
REFERENCES:
patent: 5321608 (1994-06-01), Namba et al.
patent: 5490061 (1996-02-01), Tolin et al.
patent: 5734889 (1998-03-01), Yamaguchi et al.
patent: 5761637 (1998-06-01), Chino
patent: 5799276 (1998-08-01), Komissarchik et al.
patent: 5828991 (1998-10-01), Skiena et al.
patent: 5933525 (1999-08-01), Makhoul et al.
patent: 5960385 (1999-09-01), Skiena et al.
patent: 5966686 (1999-10-01), Heidorn et al.
patent: 5991721 (1999-11-01), Asano et al.
patent: 40-2016671A (1990-01-01), None
patent: 40-5225175A (1993-09-01), None
DAILOG File 275, Acc. No. 01288595: Sheryl R. Young, et al.: “High Level Knowledge Sources in Usable Speech Recognition Systems”, Communications of the ACM, vol. 32, No. 2, pp. 183 (12), Feb. 1989.
DIALOG File 275, Acc. No. 01523808: Ellen Germain: “Introducing Natural Language Processing” (Tutorial); Al Expert, vol. 7, No. 8, pp. 30 (6), Aug. 1992.*
Abe Keiko
Katsura Koyo
Kuzunuki Soshiro
Miura Masaki
Yokota Toshimi
Antonelli Terry Stout & Kraus LLP
Hitachi , Ltd.
Thomas Joseph
LandOfFree
Sentence processing apparatus and method thereof,utilizing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Sentence processing apparatus and method thereof,utilizing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Sentence processing apparatus and method thereof,utilizing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2531142