Malayalam Text summarization Using Vector Space Model

Automatic text summarization systems extract the significant sentences from the document and generate an accurate summary. The technique of text summarization is abstractive and extractive. Abstractive summarization understands the source text and generates new shorter text with same ideas. It requires language processing tools like Dictionaries, WordNet etc. Extractive summarization systems find the semantics of sentences and rank the semantically similar sentences and high scored sentences are selected to generate a summary. In extractive summarization statistical and linguistic methods are used to rank the sentences. The high scored sentences are selected as summary. Many techniques have been developed for summarization of text in various languages. In Malayalam, summarization systems are very few and it is in the beginning stage. This paper discusses about the semantic similarity method like vector space model and shows how ranking the sentences using this model and also gives the efficiency of proposed summarizer.