Automatic Part Of Speech Tagging For Amharic Language An Experiment Using Stochastic Hidden Markov (hmm) Approach

Natural Language processing, as a field of scientific inquiry, plays an important role inrnincreasing computers capability to understand natural languages. Part of speech (POS)rntagging is one effort in the task of understanding natural language, the language by whichrnmost human knowledge is recorded. The task of POS tagging is to assign unique part of speech tags to sentences that are presentedrnas a linear string of words. POS tagging systems, which annotate corpora written in variousrnlanguages (e.g. English), are used as components in many applications including phrasernrecognition, word sense disambiguation, grammatical function assigmnents and many others.Today, taggers of different kinds have been developed for languages, which have relativelyrnwider use nationally and/or internationally. The same story is not true for Amharic, thernworking language of the Federal Government of Ethiopia, and one of the major languages ofrnEthiopia (Bender, 1976) for there are no systems (taggers of any sort) that all Notate corporarnwritten in this language.

Get Full Work

Report copyright infringement or plagiarism