A Comparative Study of Extractive Summary Algorithms Using Natural Language Processing

Abstract

Language, an important feature in our daily life. It is a tool in communication used by humans. Currently, there is an increasing number of articles and journals flooded on the Internet It is hard to read and study au the related articles to users’ research areas manually because there is limited time for each people. One of the solutions is to summarize texts in the article. Natural Language Processing (NLP) is one of the features in Machine Learning (ML) and it is used for summarization This study was tried to investigate the performances of the three different extractive algorithms from NLP. The results were evaluated with the ROUGE evaluation package using ROUGE-1, ROUGE-2, ROUGE-L, and ROUGE-SU4 methods. 3 $\theta$ samples from the BBC dataset were used as the training data in the evaluation process. Results generated from ROUGE toolkit show the performance of the Barrios et al.’s works is the best among the other two.