| US 7,548,845 B2 | ||
| Apparatus, method, and program product for translation and method of providing translation support service | ||
| Eiju Akahane, Yokohama (Japan); and Yumi Ikeza, Yokohama (Japan) | ||
| Assigned to International Business Machines Corporation, Armonk, N.Y. (US) | ||
| Filed on Dec. 08, 2006, as Appl. No. 11/608,368. | ||
| Claims priority of application No. 2005-355001 (JP), filed on Dec. 08, 2005. | ||
| Prior Publication US 2007/0150259 A1, Jun. 28, 2007 | ||
| Int. Cl. G06F 17/28 (2006.01); G06F 17/21 (2006.01) | ||
| U.S. Cl. 704—7 [704/2; 704/10] | 3 Claims |
| 1. A method for providing a translation support service by using an apparatus for selecting from a translation record for
use in translation of a newly received text among a plurality of translation records, each of the translation records storing
with respect to past translation results at least one pair of a source-language text, being a divided part corresponding to
a translation segment in a received unit, being units of translation of text, and a target-language text corresponding to
the source-language text, the method comprising:
a first key generation step of, with respect to each of the translation segments in the received text, generating an input
key in which the source-language text, being a divided part corresponding to one of the translation segments, is encoded based
on a predetermined conversion rule by a first key generation unit in a computer;
an acquisition step of, in association with each of the plurality of translation records, acquiring a translation record key
in which the source-language text, being the divided part corresponding to one of the translation segments in the translation
record, is encoded based on the predetermined conversion rule by an acquisition unit in the computer;
a key search step of searching whether the translation record key identical to each of the input keys is present by a key
search unit in the computer, with respect to each of the translation records;
a first count step of counting the quantity of input keys that have translation record keys identical therewith by a first
count unit realized by the computer, with respect to each of the translation records; and
a determination step of determining the translation record for use in translation of the received text by a determination
unit realized by the computer, on the basis of a count result with respect to each of the translation records;
wherein the input key generated by first key generation step comprises a character string in which a determined number of
characters is selected from the leading end of each word contained in a source-language text, being a divided part corresponding
to one of translation segments in the received text, and concatenated, and
wherein the translation record key generated by the acquisition step, in association with each of the translation records,
comprises a character string in which a determined number of characters is selected from the leading end of each word contained
in the source-language text, being a divided part corresponding to one of the translation segments in the translation record,
and concatenated.
|