Cloud Translation System Based on Hadoop and XMPP
YU Zheng-hon · Computer and Digital Engineering · 2014
Traditional machine translation systems have disadvantages such as poor accuracy,labor cost and so on.This paper proposed translation system solutions based on cloud computing Hadoop framework and XMPP protocol,combining traditional translation technology and cloud computing Hadoop framework which used XMPP interoperability among heterogeneous systems,created a user,the interpreter and the object of the tripartite mutual cloud platform.The cloud translation system can be mutual communication process mining complex corpus of resources.It has several brilliant features,such as a large corpus of data,accurate translation,translation efficiency,intelligence and so on.By using this system we solved the high cost of manual translation,machine translation ambiguity and other issues,to achieve a different language groups for text instant messaging via the Internet when multilingual communication barrier.