Topic-based Amharic Text Summarization

Automatic text summarization is important in today’s information age where vast amount ofrninformation are produced for consumption. The case of Ethiopia is not an exception. The country hasrnseen steady growth in digital content, ready for consumption by the mass. Compared to otherrninternational languages, text summarization works in Ethiopia’s local languages in general and thernAmharic language in particular, can be said to be in their early stages of development. In this regard,rnmore work should be carried out to meet present and future needs of the availability of high qualityrninformation that needs to be extracted from large collections of data in a timely manner.rnThis thesis investigates the problem of building a concept-based single-document Amharic textrnsummarization system. Because local languages like Amharic lack extensive linguistic resources, wernpropose to use statistical approaches called topic modeling to create our text summarizer. Thernproposed algorithms are language and domain independent and hence can also be used for other localrnlanguages. More specifically, we propose to use the topic modeling approach of probabilistic latentrnsemantic analysis (PLSA).rnWe show that a principled use of the term by concept matrix that results from a PLSA model canrnhelp produce summaries that capture the main topics of a document. We propose six algorithms tornhelp explore the use of the term by concept matrix. All of the algorithms have two common steps. Inrnthe first step, keywords of the document are selected using the term by concept matrix. In the secondrnstep, sentences that best contain the keywords are selected for inclusion in the summary. To takernadvantage of the kind of texts we experiment with (news articles) the algorithms always select thernfirst sentence of the document for inclusion in the summary.rnWe evaluated the proposed algorithms for precision/recall for summaries of 20%, 25% and 30%rnextraction rates. The best results achieved are as follows: 0.45511 at 20%, 0.48499 at 25% andrn0.52012 at 30%. We also compared our systems with previous summarization methods that havernbeen developed for other languages based on topic modeling approaches using our Amharic data set.rnOur results show that the proposed algorithms perform better at all extraction rates.rnKeywords: Amharic Text Summarization, Keyword Approach, Probabilistic Latent Semantic Analysis

Get Full Work

Report copyright infringement or plagiarism