Text Summarization Using Text Frequency Ranking Sentence Prediction

In the era of information technology, data plays significant role. The data which prevails on the internet are unstructured and are not in a concise manner. To make the raw data into a structured, readable, coherent and concise and to extract the summary of data, the text summarization concept is introduced. The text summarization involves in providing a summary of the useful information from the raw data without dissolving the main theme of the data. Nowadays readers face the challenge of reading comments, reviews, news articles, blogs, etc., as they are too informal and noisy. Retrieving the correct gist of the text which is necessary for all the readers is a quite difficult task. In order to overcome the problems faced by the readers, TFRSP (Text Frequency Ranking Sentence Prediction) algorithm is proposed to generate a precise summary that uses supervised and unsupervised learning algorithms. The proposed approach uses the combination of TF-IDF-TR (Term Frequency - Inverse Document Frequency - Text Rank) as an unsupervised learning algorithm and Seq2Seq (Sequence to Sequence) model as a supervised learning algorithm to obtain the benefits of both extractive and abstractive summarization. The results of the proposed TFRSP approach is compared with the existing methods of text summarization using the Recall Oriented Understudy of Gisting Evaluation (ROUGE) and attains a high ROUGE score, hence achieves high accuracy of summary.