Efficient feature integration with Wikipedia-based semantic feature extraction for Turkish text summarization

Abstract

This study presents a novel hybrid Turkish text summarization system that combines structural and semantic features. The system uses 5 structural features, 1 of which is newly proposed and 3 are semantic features whose values are extracted from Turkish Wikipedia links. The features are combined using the weights calculated by 2 novel approaches. The first approach makes use of an analytical hierarchical process, which depends on a series of expert judgments based on pairwise comparisons of the features. The second approach makes use of the artificial bee colony algorithm for automatically determining the weights of the features. To confirm the significance of the proposed hybrid system, its performance is evaluated on a new Turkish corpus that contains 110 documents and 3 human-generated extractive summary corpora. The experimental results show that exploiting all of the features by combining them results in a better performance than exploiting each feature individually.