Automatic Text Summerization

Information Technology Engineering

Extractive Text Summarization is a Natural Language Processing problem of forming a summary by using the most important sentences of an article and has been addressed in many different ways. In this research, a model is developed that uses metrics for giving importance to sentences, and applies it to Machine Learning and Deep Learning Models. The main objective of the model is to learn to classify which sentences belong and which do not in a summary. Using the results obtained the optimal approach, along with other approaches that work particularly well, to do extractive text summarization, are found. In addition, the analysis of why these models perform better than the rest is done. Finally, the importance of each metric in forming a summary is discovered based on several evaluation measures. Text summarization has a lot of research devoted to it, which has resulted in development of various techniques to do the same. The process of text summarization is broadly divided into two types – extractive and abstractive. To develop a model that does extractive text summarization along with finding and evaluating which parameters and machine learning and deep learning algorithms optimize the working of the model.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.