Hybrid auto text summarization using deep neural network and fuzzy logic system

Amount of text for a particular topic is increasing at an exponential rate each day, causing availability of bulk text data, some of which will be relevant to our search while some data may be irrelevant. In such case we need to summarize the text document, so as to understand the key elements of the document. A Hybrid Automatic Summarizer using soft computing techniques namely, Fuzzy logic system and deep neural system is proposed. The summary is based on extractive summarization technique. Restricted boltzman Machine (RBM) is used in the deep neural network and fuzzy rule base on the sentences for feature extraction using a sentence matrix. Evaluation of hybridized systems gave better summary as compared to the individual systems.