E-services translation utilizing machine translation and translation memory
View Patent ↗A system and method for translating data from a source language to a target language is provided wherein machine generated target translation of a source sentence is compared to a database of human generated target sentences. If a matching human generated target sentence is found, the human generated target sentence may be used instead of the machine generated sentence, since the human generated target sentence is more likely to be a well-formed sentence than the machine generated sentence. The system and method does not rely on a translation memory containing pairs of sentences in both source and target languages, and minimizes the reliance on a human translator to correct a translation generated by machine translation.
1. A method for translating data from a source language to a target via one or more computing systems, comprising:
inputting source data for translation;
searching via a processor a database for a source string similar to the inputted source data, said source string having a corresponding translated target string;
outputting said corresponding translated target string, if said source data string similar to said inputted source data string is found, for subsequent review by a translator;
machine translating said inputted source data if said source data string similar to said inputted source data string is not found, said machine translating producing a target string;
searching via the processor said database for a target string matching the target string produced by said machine translating;
sending said matching target string for subsequent review by said translator if said matching target string is found; and
sending said machine translated target string for subsequent review by said translator if a match for said machine translated target string is not found in said database.
2. The method as recited in claim 1 , wherein said database comprises strings of source data and matching strings of target data.
3. The method as recited in claim 2 , wherein said database further comprises unmatched strings of source data and strings of target data.
4. The method as recited in claim 1 , wherein said machine translating uses units of data smaller than a sentence.
5. The method as recited in claim 1 , wherein said machine translating uses a bilingual corpus to align sentences, phrases and words from said source language to said target language.
6. The method as recited in claim 1 , wherein said database comprises strings of source data and strings of target data, said strings of source data and strings of target data being unmatched.
7. The method as recited in claim 1 , wherein said database comprises a collection of source language sentences and target language sentences, where none of said source language sentences and said target language sentences are paired together.
8. A method for translating data from a source language to a target language, comprising:
inputting source data for translation;
searching via a processor a database for a source string similar to the inputted source data, said source string having a corresponding translated target string, said database comprising matching strings of source data and strings of target data and unmatched strings of source data and strings of target data;
outputting said corresponding translated target string, if said source data string similar to said inputted source data string is found, for subsequent review by a translator;
machine translating said inputted source data if said source data string similar to said inputted source data string is not found, said machine translating producing a target string, wherein said machine translating uses units of data smaller than a sentence;
searching via said processor said database for a target string matching the target string produced by said machine translating;
sending said matching target string for subsequent review by said translator if said matching target string is found; and
sending said machine translated target string for subsequent review by said translator if a match for said machine translated target string is not found in said database.
9. The method as recited in claim 8 , wherein said machine translating uses a bilingual corpus to align not sentences, phrases and words from said source language to said target language.
10. A system for translating data from a source language to a target language, comprising:
a device to input source language data to be translated;
a translation database comprising source language string and target language strings;
a translation source connected to a network; and,
machine translator that translates said input source language data into a target language string;
a portal for searching said translation database for a target string matching the target language string produced by said machine translating, and sending said matching target language string for subsequent review by said translation source if said matching target string from said translation database is found, and sending said machine translated target string from said machine translator for subsequent review by said translation source if said matching target string from said translation database is not found in said translation database.
11. The system as recited in claim 10 , wherein said machine translator uses units of data smaller than a sentence.
12. The system as recited in claim 10 , wherein said machine translator uses a bilingual corpus to align sentences, phrases and words from said source language to said target language.
13. The system as recited in claim 10 , wherein said translation database comprises strings of source data and strings of target data, said strings of source data and strings of target data being unmatched.
14. The system as recited in claim 10 , wherein said translation database further comprises unmatched strings of source data and strings of target data.
15. The system as recited in claim 10 , wherein said translation database comprises a collection of source language sentences and target language sentences, where none of said source language sentences and said target language sentences are paired together.