Opinion Summarisation using Bi-Directional Long-Short Term Memory

This generation, people depend on multiple resources to stay up-to-date whether be it social media or a new product available on the market. To be able to develop a model that can effectively summarise a long paragraph or a long comment posted on a website can be useful to grasp the primary information provided in that article or review hence reducing the time input to make a decision or to come to an understanding of the data. In this paper we make use of a deep learning model that helps us to get the summaries from the lengthy reviews provided by the customers. Summarisation is the method of getting a summary from a sequence of texts or files that helps us understand the basic content of the information within them. Opinion summarisation is the method of getting opinions from a set of sentences. It can be done through extractive and abstractive methods. Here we use abstractive summarisation that learns from the input dataset and comes through with thee summary with the best possible result. Our paper describes an abstractive method to get the summaries from a dataset using Bi-directional Long-Short Term Memory. We make use of an Attention layer to increase the performance of the model and help to improve the efficiency of the model. The model effectiveness is evaluated using ROUGE and BLUE scores.