Markov Models Applications in Natural Language Processing: A Survey

Markov models are one of the widely used techniques in machine learning to process natural language. Chains and Hidden Models stochastic employed for modeling systems that dynamic where future state relies on current state. The chain, which generates a sequence words create complete sentence, is frequently generating hidden model named-entity recognition tagging parts speech, tries predict tags based observed words. This paper reviews models' use three applications language processing (NLP): generation, recognition, speech tagging. Nowadays, researchers try reduce dependence lexicon or annotation tasks NLP. In this paper, we have focused as approach A literature review was conducted summarize research attempts with focusing methods/techniques NLP, their advantages, disadvantages. Most NLP studies apply supervised improvement using decrease dependency tasks. Some others unsupervised solutions reducing labeled datasets.