Name Entity Recognition for Myanmar Language

Abstract

Name Entity Recognition (NER) is one of the fundamental problems in many Natural Language Processing applications. NE consists of names, person, location, organization, time, numerical expressions and etc. Not much works have been done on NER for Myanmar language. Due to the characteristics of Myanmar language, NE extraction is more difficult than English Language. Myanmar Language is also difficult to distinguish NE and common word because it lacks of orthographical information to signal NEs, no boundary indication between words and no capitalization like English. In this paper, we propose NER algorithm which can recognize three classes of NEs: Person, Organization and Location. In identification phase, we employed pre-defined rules generated from Language knowledge, and Myanmar-English bi-lingual Lexicon.