Automatic Mining of Verb Affixes for Myanmar Language with Unsupervised learning by using N-Grams

Abstract

A separable verb is a verb that is composed of a verb stem and a separable affix which is called verb affixes. Mining of verb affixes is to extract the various forms of verb affixes from any Myanmar sentences. A Method for mining verb affixes in Myanmar language is presented in this paper. An unsupervised learning approach is used to extract affixes for various forms of verbs from Myanmar sentences since the verb affixes are supported to restrict verb from any Myanmar sentences. Myanmar texts sentences are analyzed to mine various forms of verbs. Unsupervised learning is the task of learning without labeled data in NLP application. Since Myanmar sentences are written with no special delimiter such as space to indicate word boundary, merging based syllable segmentation method is used to indicate word boundary. Verb affixes are extracted from these sentences by using N-Grams method. The system is implemented using java.