Automatic Construction Of Amharic Semantic Networks (asnet)

Semantic networks are becoming popular issues these days. Even though this popularity isrnmostly related to the idea of semantic web, it is also related to the natural language applications.rnSemantic networks allow search engines to search nOI only for the key words given by the userrnbut also for the related concepts. and show how this relation is made. Knowledge stored asrnsemantic networks can he used by programs that generate text from structured data. Semanticrnnetworks are also used for document summarization by compressing the data semantically andrndocument classification using the knowledge stored in it. As a result, semantic networks havernbecome key components in many NLP applications.rnIn this thesis, we focused on the construction of semantic networks for Amharic text. We haverndeveloped Amharic WordNet as initial knowledge base for the system and extracted interveningrnword patterns between pairs of concepts in the WordNet for a specific relation from free text.rnFor each pair of concepts which we know the relationship contained in Amharic WordNet, wernsearch the corpus for some text snapshot between these concepts. The returned text snapshot isrnprocessed to extract all the patterns having n·gram words between the two concepts. We havernused the WordS pace model for extraction of semantically related concepts. The process ofrnrelation identification in among these concepts utilizes the extracted text patterns. "Part·of' andrn"type·of' relations are very popular and frequently found between concepts in any corpus. Wernhave designed our system to extract "part·of' and "type·of' relations between concepts.rnThe system was tested in three different phases with different datasets from Ethiopian NewsrnAgency and Walta Information Center. The accuracy of the system to extract pairs of conceptsrnhaving "type·of' and "part-of' relations is 68.5% and 71.7% respectively.

Get Full Work

Report copyright infringement or plagiarism