Scholarlink

This paper describes the methodology of our Corpus-Based Machine Translation System using Word Embeddings for English–Hindi language pair. We used the parallel corpus developed at Banasthali Vidyapith and trained a word2vec model on it. We then created a vector database for the corpus which had cross-lingual word vectors. These were used in searching and matching the source sentence and finally were translated into the target text. We evaluated the system using the standard MT evaluation metrics which produced satisfactory results.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Implications of Word Embeddings on Corpus-Based Machine Translation

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Martina David,
  • Nisheeth Joshi

摘要

This paper describes the methodology of our Corpus-Based Machine Translation System using Word Embeddings for English–Hindi language pair. We used the parallel corpus developed at Banasthali Vidyapith and trained a word2vec model on it. We then created a vector database for the corpus which had cross-lingual word vectors. These were used in searching and matching the source sentence and finally were translated into the target text. We evaluated the system using the standard MT evaluation metrics which produced satisfactory results.