Automatic Text Summarization: Soft Computing Based Approaches

In this fast moving world, digital publications such as online text articles or documents can be easily searched and obtained via the Internet. However, the most important issue is how to digest this information and provide them to the readers in a concise form. Many researchers have conducted studies on automatic text summarization and various approaches have been proposed to improve the quality of summary. In this paper, a study is conducted on how soft computing based approaches have been used to improve automatic text summarization. The commonly used soft computing approaches in text summarization are discussed, namely neural network based method, fuzzy logic based method, evolutionary computing based method and Bayesian statistics based method. Besides the general idea and concepts of each method, the benefits and limitations concerning these methods are identified. At the end of this paper, a hybrid soft computing approach for text summarization is proposed.