Preprocessing, term selection, term weighting, sentence weighting, and sentence selection are the main issues in generating extractive summaries of text sentences. Although many outstanding related works only are focused in the last step, they show sophisticated features in each one. In order to determine the relevance of the sentences (sentence selection step) many sentence features have been proposed in this task (in fact, these features are related to all the steps). Recently, some good related works have coincided in the same features but they present different ways for weighting these features. In this paper, a method to optimize the combination of previous relevant features in each step based on a genetic algorithm is presented. The proposed method not only outperforms previous related works in two standard document collections, but also shows the relevance of these features to this problem.

The main problem for generating an extractive automatic text summary is to detect the most relevant information in the source document. For such purpose, recently some approaches have successfully employed the word sequence information from the self-text for detecting the candidate text fragments for composing the summary. In this paper, we employ the so-called n-grams and maximal frequent word sequences as features in a vector space model in order to determine the advantages and disadvantages for extractive text summarization.