Hybrid translation apparatus and method thereof

1. A hybrid translation apparatus comprising:

a computer system, comprising:

a source language input unit for generalizing an input source language sentence for each node;

a statistics-based translation knowledge database(DB) for storing learning data generalized for each node to be acquired to utilize the learning data as transformation knowledge for statistics-based translation;

a first translation result generating unit for transforming the source language sentence generalized for each node into a node expression using the statistics-based translation knowledge to generate a first translation result;

a second translation result generating unit for repeatedly performing the generation of a target word for each node on the first translation result using pattern-based knowledge to generate a second translation result as target words for the respective nodes;

a pattern-based translation knowledge DB for using in the pattern-based translation; and

a translation result combining unit for combining the first translation result and the second translation result to output a target language sentence,

wherein the source language input unit performs the generic transformation for a node expression using all sentence pairs of a pre-constructed parallel corpus, and stores acquired training data in a database of the statistics-based translation knowledge DB.

2. The apparatus of claim 1 , wherein the source language input unit acquires syntactic information and a syntactic range for identifying the lexicon and part-of-speech for each morpheme in order to transform a translation node in the source language sentence by language analysis.

3. The apparatus of claim 1 , wherein the first translation result generating unit collects translation information based on the statistical relationship between an input sentence and a translated sentence to generate the first translation result based on the statistics-based translation knowledge DB for each input.

4. The apparatus of claim 1 , wherein the second translation result generating unit repeatedly performs the generation of a target word for each node using the pattern-based translation knowledge including a bilingual dictionary and a translation pattern dictionary until the translation of all nodes is complete.

5. The apparatus of claim 4 , wherein the second translation result generating unit generates the second translation result using a translation dictionary and pattern information described in syntactic units for the transformation of a source language and a translated sentence.

6. The apparatus of claim 1 , wherein the translation result combining unit performs combination by inserting a target word for each node based on the second translation result as a substitute to each node position of the first translation result.

7. The apparatus of claim 6 , wherein the translation result combining unit generates the target language sentence by refining the connection of the sentence with respect to a combination result of the first translation result and the second translation result.

8. A hybrid translation method comprising:

generalizing an input source language sentence for each node;

transforming, using a computer, the source language sentence generalized for each node into a node expression using statistics-based translation knowledge DB to generate a first translation result;

store learning data generalized for each node to be acquired to utilize the learning data as transformation knowledge for statistics-based translation;

repeatedly performing the generation of a target word for each node on the first translation result using pattern-based knowledge to generate a second translation result as target words for the respective nodes; and

combining the first translation result and the second translation result to output a target language sentence,

wherein before said generalizing:

performing the generic transformation for a node expression using all sentence pairs of a pre-constructed parallel corpus, and storing acquired training data in a database of the statistics-based translation knowledge DB.

9. The method of claim 8 , wherein said combining performs combination by inserting a target word for each node based on the second translation result as a substitute to each node position of the first translation result.

10. The method of claim 9 , wherein said combining generates the target language sentence by refining the connection of the sentence with respect to a combination result of the first translation result and the second translation result.

11. The method of claim 8 , wherein said generalizing acquires syntactic information and a syntactic range for identifying the lexicon and part-of-speech for each morpheme in order to transform a translation node in the source language sentence by language analysis.

12. The method of claim 8 , wherein said transforming collects translation information based on the statistical relationship between an input sentence and a translated sentence to generate the first translation result based on the statistics-based translation knowledge DB for each input.

13. The method of claim 8 , wherein said repeatedly performing repeatedly performs the generation of a target word for each node using the pattern-based translation knowledge including a bilingual dictionary and a translation pattern dictionary until the translation of all nodes is complete.

14. The method of claim 13 , wherein said repeatedly performing generates the second translation result using a translation dictionary and pattern information described in syntactic units for the transformation of a source language and a translated sentence.