Word as an agent for multi-agent based machine translation

Abstract

Machine translation systems use various approaches including human-assisted, rule-based, example-based, statistical and multi-agent. The multi-agent approach is a modern approach to handle complexity of the systems in past five years. This multi-agent approach has been successfully used to develop machine translation systems by using language primitives (words) as agents. Words are the basic language primitives in any language and these language primitives on involved in major rules to represent morphology, syntax and semantics. Therefore, word agents are used to handle syntax and semantic in many multi-agent based machine translation. The ontology of the single word agent represents morphology of the particular word and group of word agents represent syntax and these information are used to identify the semantics and pragmatics of the source text. This paper presents overview of the Multi-agent based English to Sinhala machine translation system. Multiagent approach for machine translation, morphology, syntax and semantic handling techniques, tasks of the word agent have been discussed in the paper.