Translation apparatus and method
AssigneeKabushiki Kaisha Toshiba
InventorKenji Ono
Life of this patent
Estimating the term…
Abstract
A translation apparatus translates article information in a first language including an article body and a related headline as a summary of the article body. A decision unit discriminately identifies the article body and the headline in the article information. A translation unit respectively translates the article body and the headline into a second language based on identification result of the decision unit.
Claims25
A translation apparatus for translating article information in a first language including an article body and a related headline into a second language, comprising:
- a decision unit configured to electronically obtain the article information, and to discriminately identify the article body and the headline in the article information;
- a phrase alignment processing unit configured to detect continuous capital letters as an abbreviation from the headline, to detect continuous words each including an initial capital letter as a formal phrase from the head sentence of the article body, to compare the continuous capital letters with a set of combined capital letters extracted from the continuous words, to count as one coincident word if the continuous capital letters coincide with the combined capital letters, and to generate the continuous capital letters and the continuous words as a phrase alignment result;
- and
- a translation unit configured to translate the article body and the headline into the second language based on the continuous capital letters and the continuous words in the phrase alignment result.
A translation method for translating article information in a first language including an article body and a related headline into a second language, comprising:
- electronically obtaining the article information;
- discriminately identifying the article body and the headline in the article information;
- detecting continuous capital letters as an abbreviation from the headline;
- detecting continuous words each including an initial capital letter as a formal phrase from the head sentence of the article body;
- comparing the continuous capital letters with a set of combined capital letters extracted from the continuous words;
- counting as one coincident word if the continuous capital letters coincide with the combined capital letters;
- generating the continuous capital letters and the continuous words as a phrase alignment result;
- and
- translating the article body and the headline into the second language based on the continuous capital letters and the continuous words in the phrase alignment result.
A computer program product, comprising:
- a computer readable program code embodied in said product for causing a computer to translate article information in a first language including an article body and a related headline into a second language, said computer readable program code having:
- a first program code to electronically obtain the article information;
- a second program code to discriminately identify the article body and the headline in the article information;
- a third program code to detect continuous capital letters as an abbreviation from the headline;
- a fourth program code to detect continuous words each including an initial capital letter as a formal phrase from the head sentence of the article body;
- a fifth program code to compare the continuous capital letters with a set of combined capital letters extracted from the continuous words;
- a sixth program code to count as one coincident word if the continuous capital letters coincide with the combined capital letters;
- a seventh program code to generate the continuous capital letters and the continuous words as a phrase alignment result;
- and
- an eighth program code to translate the article body and the headline into the continuous capital letters and the continuous words in the second language based on the phrase alignment result.
Dependencies are inferred from the claim text.