Improve Extractive Voting-Based Automated Text Summarization using two-step ranking

Automatic Text Summarization (ATS) can be defined as using computer programs to reduce the size of a text while preserving its information content. And a selection of some sentences of the original document as the generated summary is extractive summarization. In this paper, an extractive voting-based text summarization method is proposed with a new approach for sentence ranking we called it two-step ranking. Initially, each sentence from the original text is scored based on each feature and then votes of all features will be used to obtain a total score of the sentence. Then sentences will be ranked using this score and top ranks will be chosen as the final summary. The proposed method was evaluated with rouge standard using DUC dataset. F-measure in this method improved 3% in rouge-1 and 5% in rouge-2 (maximum value).