Pdtb Style Sentence Level Shallow Discourse Parser For Amharic

Research on natural language processing applications (NLP) is a very important topic in our dailyrnlife, by enabling computers to understand human languages. Such researches has come a long wayrnin foreign languages like English, Japanese, Chinese, Portuguese and Arabic. NLP applicationsrnsuch as include machine translation, question answering, knowledge extraction and informationrnretrieval are some of the fruits of such researches.rnDiscourse parser is one of the main components that enables the realization of such NLPrnapplications. For foreign languages like English and Arabic, many discourse parsers are developedrnin different approaches. However, in the case of Amharic, there are no works done, to the best ofrnthe researcher’s knowledge, on Amharic discourse parser so far.rnIn this study, a Penn Discourse Tree Bank (PDTB) style sentence level shallow discourse parserrnfor Amharic is developed. We have used machine-learning algorithms to accomplish the subtasksrnof discourse parsing. The algorithms utilize lexical and positional features of the discourse markerrnand related words for segmentation and identify associated discourse relation. The parser is testedrnon test sentences, which are extracted from different sources. Encouraging results are observedrnfrom the experiments performed,

Get Full Work

Report copyright infringement or plagiarism