Extractive Three-line Summarization of News Articles using Neural Network

Abstract

The purpose of text summarization is to enable readers to understand the point of documents without reading overall of them. With the spread of smartphones, opportunities to view news articles within the limited screen area are increasing, so that there are needs easy to read summaries within the limited screen area. In this paper, we propose a neural network model that extracts three-line summaries from news articles. As a dataset, we use articles and three-line summaries of Livedoor NEWS on the Internet. In an experiment, we show that the proposed model performs better than a conventional method.