Abstract
Automatic Text Summarization (ATS) enables users to save their precious time to retrieve their relevant information need while searching voluminous big data. Text summaries are sensitive to scoring methods, as most of the methods requires to weight features for sentence scoring. In this chapter, various statistical features proposed by researchers for extractive automatic text summarization are explored. Features that perform well are termed as best features using ROUGE evaluation measures and used for creating feature combinations. After that, best performing feature combinations are identified. Performance evaluation of best performing feature combinations on short, medium and large size documents is also conducted using same ROUGE performance measures.
Reference56 articles.
1. Differential evolution cluster-based text summarization methods
2. Text summarization features selection method using pseudo Genetic-based model
3. A new sentence similarity measure and sentence based extractive technique for automatic text summarization
4. Baldwin, B., & Morton, T. S. (1998). Dynamic co reference based summarization. In Proceedings of the Third Conference on Empirical Methods in Natural Language Processing (EMNLP-3).
5. Using Lexical Chains for Text Summarization;R.Barzilay;Proceedings of the ACL/EACL’97 Workshop on Intelligent Scalable Text Summarization,1997