Offering a New Approach for Approximate Pattern Matching in Example-Based Machine Translation

Reza Akbari; Mohammad Hadi Sadreddini; Seyed Mostafa Fakhrahmad

In this article, a new model is proposed in order to measure the degree of similarity between two sentences in machine translation based on example. The proposed model has applied genetic algorithm beside a new fitness function which is based on semantic load matching between the two sentences. Here, verbs are considered as the heart of a sentence because they are the main part of a sentence and carry the major part of the semantic load in the sentence; therefore more attention is paid to the verbs in the fitness function. It is noteworthy that the proposed model is largely dependent on the verbal part and the extracted synonyms from WordNet as well as the arrangement of words. The results are promising by precision and recall, indicating that the proposed method improves the quality of the retrieved matched sentences.

Select Volume / Issues:

• Volume 6 Issue 1, Jan. 2015

Year:

2015

Type of Publication:

Article

Keywords:

Machine Translation; Example Based Machine Translation; Matching Phase; The Similarity of Sentences; Semantic Load; Genetic Algorithm

Journal:

IJECCE

Volume:

6

Number:

1

Pages:

131-139

Month:

January

ISSN:

2249-071X