Input Text Repairing for Multi-lingual Chat System

Even though various communication tools have resulted in a remarkable increase of global communications, language barriers remain high and complicate communication across languages. Although the multi-lingual chat system allows users to chat with each other in different language using machine translation. the quality of translation is not so high when the input sentence reflects spoken language. In this paper we propose a method that repairs the input sentences in spoken language by retrieving similar sentences using keywords.