Semantic based Text Summarization for Single Document on Android Mobile Device

The explosion of information in the World Wide Web is overwhelming readers with limitless information. Large internet articles or journals are often cumbersome to read as well as comprehend. More often than not, readers are immersed in a pool of information with limited time to assimilate all of the articles. It leads to information overload whereby readers are trying to deal with more information than they can process. Hence, there is an apparent need for an automatic text summarizer as to produce summaries quicker than humans. The text summarization research on mobile platform has been inspired by the new paradigm shift in accessing information ubiquitously at anytime and anywhere on Smartphones or smart devices. In this research, a semantic and syntactic based summarization is implemented in a text summarizer to solve the overload problem whilst providing a more coherent summary. Additionally, WordNet is used as the lexical database to semantically extract the text document which provides a more efficient and accurate algorithm than the existing summary system. The objective of the paper is to integrate WordNet into the proposed system called TextSumIt which condenses lengthy documents into shorter summarized text that gives a higher readability to Android mobile users. The experimental results are done using recall, precision and F-Score to evaluate on the summary output, in comparison with the existing automated summarizer. Human-generated summaries from Document Understanding Conference (DUC) are taken as the reference summaries for the evaluation. The evaluation of experimental results shows satisfactory results.