Translation support system, etc.

A translation support device for supporting machine translation from a source-language sentence into a target-language sentence, the translation support device comprising:

• an input unit that accepts an input of a source-language sentence to be translated;
• an error database that at least stores words or combinations of words included in a plurality of source-language sentences for which machine translation from the source-language sentences into target-language sentences is not performed correctly;
• a controlled-source-language-sentence corpus that stores a plurality of source-language sentences as well as controlled source-language sentences, which are source-language sentences that are controlled, corresponding to the plurality of source-language sentences and expressed in a format satisfying predetermined conditions;
• a control unit that classifies whether or not the input source-language sentence is machine-translatable;
• and
• an output unit that is capable of outputting the input source-language sentence classified as being non-machine-translatable,
• wherein the control unit:
• determines a score representing the complexity of the input source-language sentence on the basis of the sentence structure of the input source-language sentence, classifies whether or not the input source-language sentence is machine-translatable according to the level of the score, and in the case where the input source-language sentence is classified as being machine-translatable, further classifies whether or not the input source-language sentence is machine-translatable on the basis of a word or a combination of words included in the input source-language sentence as well as the words or combinations of words stored in the error database;
• and
• in the case where the input source-language sentence is classified as being non-machine-translatable, classifies whether or not the input source-language sentence is convertible into the controlled-source-language sentence on the basis of the degrees of matching between the word or the combination of words included in the input source-language sentence and the words or the combination of words included in the source-language sentences stored in the controlled-source-language-sentence corpus, and in the case where the input source-language sentence is classified as being convertible, converts the input source-language sentence into the controlled-source-language sentence by using the controlled-source-language-sentence corpus and classifies the input source-language sentence as being machine-translatable, and in the case where the input source-language sentence is classified as being non-convertible, classifies the input source-language sentence as being non-machine-translatable.

A non-transitory computer readable medium storing a program for supporting machine translation from a source-language sentence into a target-language sentence, the program causing a computer to execute:

• an input accepting step of accepting an input of a source-language sentence to be translated;
• a first classifying step of determining a score representing the complexity of the input source-language sentence on the basis of the sentence structure of the input source-language sentence and classifying whether or not the input source-language sentence is machine-translatable according to the level of the score;
• a second classifying step of classifying, in the case where the input source-language sentence is classified as being machine-translatable in the first classifying step, whether or not the input source-language sentence is machine-translatable on the basis of a word or a combination of words included in the input source-language sentence as well as words or combinations of words stored in a database that at least includes words or combinations of words included in a plurality of source-language sentences for which machine translation from the source-language sentences into target-language sentences is not performed correctly;
• a third classifying step of classifying, in the case where the input source-language sentence is classified as being non-machine-translatable in the first classifying step or the second classifying step, whether or not the input source-language sentence is convertible into the controlled-source-language sentence on the basis of the degrees of matching between the word or the combination of words included in the input source-language sentence and words or combinations of words included in source-language sentences stored in a corpus that stores a plurality of source-language sentences as well as controlled source-language sentences, which are source-language sentences that are controlled, corresponding to the plurality of source-language sentences and expressed in a format satisfying predetermined conditions, and in the case where the input source-language sentence is classified as being convertible, converting the input source-language sentence into the controlled-source-language sentence by using the corpus and classifying the input source-language sentence as being machine-translatable, and in the case where the input source-language sentence is classified as being non-convertible, classifying the input source-language sentence as being non-machine-translatable;
• and
• an output step of outputting the input source-language sentence classified as being non-machine-translatable in the third classifying step.

A method for supporting machine translation from a source-language sentence into a target-language sentence, the method being executed by a computer and comprising:

• an input accepting step of accepting an input of a source-language sentence to be translated;
• a first classifying step of determining a score representing the complexity of the input source-language sentence on the basis of the sentence structure of the input source-language sentence and classifying whether or not the input source-language sentence is machine-translatable according to the level of the score;
• a second classifying step of classifying, in the case where the input source-language sentence is classified as being machine-translatable in the first classifying step, whether or not the input source-language sentence is machine-translatable on the basis of a word or a combination of words included in the input source-language sentence as well as words or combinations of words stored in a database that at least includes words or combinations of words included in a plurality of source-language sentences for which machine translation from the source-language sentences into target-language sentences is not performed correctly;
• a third classifying step of classifying, in the case where the input source-language sentence is classified as being non-machine-translatable in the first classifying step or the second classifying step, whether or not the input source-language sentence is convertible into the controlled-source-language sentence on the basis of the degrees of matching between the word or the combination of words included in the input source-language sentence and words or combinations of words included in source-language sentences stored in a corpus that stores a plurality of source-language sentences as well as controlled source-language sentences, which are source-language sentences that are controlled, corresponding to the plurality of source-language sentences and expressed in a format satisfying predetermined conditions, and in the case where the input source-language sentence is classified as being convertible, converting the input source-language sentence into the controlled-source-language sentence by using the corpus and classifying the input source-language sentence as being machine-translatable, and in the case where the input source-language sentence is classified as being non-convertible, classifying the input source-language sentence as being non-machine-translatable;
• and
• an output step of outputting the input source-language sentence classified as being non-machine-translatable in the third classifying step.