Translation machine having capability of registering idioms

• US 5,541,838 A
• Filed: 10/25/1993
• Issued: 07/30/1996
• Est. Priority Date: 10/26/1992
• Status: Expired due to Term

• Current Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Original Assignee: Sharp Kabushiki Kaisha (Hon Hai Precision Industry Co., Ltd.)
• Inventors: Koyama, Noriyuki, Suzuki, Hitoshi, Sata, Ichiko, Fukumochi, Yoji
• Primary Examiner(s): Hayes, Gail O.
• Assistant Examiner(s): Thomas, Joseph
• Application Number:
  US08/142,778
• Time in Patent Office: 1,009 Days
• Field of Search: 364/419.02, 364/419.03, 364/419.04, 364/419.05, 364/419.08, 364/419.09, 395/700
• US Class Current: 704/4
• CPC Class Codes: G06F 40/55 Rule-based translation

Abstract

A translation machine is provided with an idiom registering capability. If a header standing for a idiom contains two or more variable parts, these parts are represented by representative symbols. With this representing method, the idiom is allowed to be easily registered and retrieved. This makes it possible to shorten the registering time, reduce the translating toil and prevent increase of information storage capacity. The translation machine includes an input unit, a storing unit, an output unit, an idiom registering unit, and a translating unit for performing a routine translating operation. The idiom registering unit operates to register an idiom header with two or more representative symbols standing for words or word trains sharing a predetermined attribute.

Claims

1. A translation machine comprising:

input means for inputting a character string and a symbol;

means for registering a header word with a plurality of symbols in a compound format and a translated counterpart of an idiom, when said idiom includes a plurality of variable parts or a variable part having a plurality of attributes, each of said symbols representing a predetermined attribute of a variable part of an idiom;

means for storing an idiom, a dictionary required for the translating process, and a processed result;

dictionary lookup and morphological analyzing means for breaking an inputted word train into morphemes and grammatically analyzing said morphemes;

idiom translating means for identifying an inputted character string, or part thereof and its attribute with the registered header word and the symbols of the idiom and generating a translated word corresponding to the identified idiom;

syntactic analyzing means for analyzing syntax of an inputted sentence of a source language;

syntax transforming means for transforming said analyzed syntax of a source language into the syntax of a target language;

means for generating a translated counterpart based on said syntax of a target language; and

output means for outputting the generated translated counterpart.

2. A translation machine as claimed in claim 1, wherein a representative symbol for a center word standing for a part of speech of an overall idiom is registered when an idiom includes a plurality of variable parts or a variable part having a plurality of attributes.

3. A translation machine as claimed in claim 1, wherein said storage means is composed of a dictionary memory having grammatical rules for translating an inputted translated sentence and information for translated words, a buffer memory for storing the process up to the creation of a translated word, and an idiom registering memory for storing an idiom registered by the idiom registering means.

4. A translation machine as claimed in claim 1, wherein said dictionary lookup and morphological analyzing means is composed of a part of speech extracting unit for breaking an inputted character string and generating information about a part of speech of each word and a translated word extracting unit for generating a candidate of a translated word of each word.

5. A translation machine as claimed in claim 1, wherein said idiom translating means is composed of an idiom retrieving unit for retrieving the idiom registering memory and selecting a header word candidate of the idiom to be matched in the representative format to the broken word train, an idiom identifying part for specifying the idiom candidate which has the attribute of a representative symbol matched to the attribute of a word or a word train corresponding to the location of the representative symbol in the idiom, an idiom analyzing unit for analyzing a syntax of a word or a word train corresponding to the representative symbol and generating a syntax of an overall idiom, and an idiom translated word generating unit for generating a translated word of the idiom of an inputted word train based on the syntax of the idiom.

6. A translation machine comprising;

means for registering a plurality of groups of words or word trains each sharing a certain feature in a classifying table;

means for registering a header having a plurality of symbols in a compound format and a translated word of an idiom with a special symbol standing for one of said groups when said idiom includes a plurality of variable parts or a variable part having a plurality of attributes, each of said special symbols representing a predetermined attribute of a variable part of an idiom;

means for identifying a word train of an inputted original sentence with said registered header of the idiom; and

means for outputting a translated word corresponding to said identified header of the idiom.

7. A translation machine provided with an idiom registering unit, said idiom registering unit comprising:

first registering means for registering a group of idioms each having a similar structure with a header having a plurality of symbols in a compound format and parallel symbols when said group of idioms includes a plurality of variable parts or a variable part having a plurality of attributes, each of said special symbols representing a predetermined attribute of a variable part of an idiom;

first identifying means for identifying an idiom contained in an inputted original sentence with said header;

output means for outputting a translated word for a matched word in said parallel symbols when the translated word for an idiom identified by said first identifying means is written by said parallel symbols;

second registering means for registering an idiom containing a predetermined variable word; and

second identifying means for identifying an idiom registered by said second registering means with said inputted sentence.