Abstract
Due to increasing amount of text data available in WWW, it becomes time consuming for information system users to explore every text source in detail. Automatic text summarization (ATS) is the process of generating summary by condensing text document automatically by a computer machine that can save users precious time. Major issue with most of the feature-based ATS methods is to find optimal feature weights for sentence scoring to optimize quality of text summary. This paper presents a novel voting-based approach that use modified reciprocal ranking approach which alleviates the issue of feature weighting and. Proposed approach use a specific prominent set of features for initial ranking that further boosts the performance. Experimental results on DUC 2002 dataset using ROUGE evaluation matrices show that our proposed voting approach performs better when compared to other statistical- and voting-based methods.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
