Paper Title
Topic Modelling with News Summary
Abstract
A huge amount of data is been collected on daily-basis. Many times, it becomes difficult to find the exact content
what we are looking for. This arises the need for some advanced tools and techniques which can search and organize the
data, thus help to understand the information.It is very important to read news or be aware of what is happening in and
around. But the problem is that in news articles we have a lot of paragraphs which feels very bored to the students or to
many people. Getting a gist or a summarized view of the news would be very helpful for them. Also, many a times people
want to see news of a specific domain like sports, films, politics, education, etc. but in news articles we see all news
together. It would have been better to give a topic and assign the summarized news under each topic. Similarly for any ECommerce
shopping portals, where people want to see the reviews of products where instead of reading all reviews serially,
one can select a topic and only the reviews related to that topic would be displayed. The purpose of this paper is to propose a
strategy of solving the issue of reading huge text and to get the exact topic content. It uses Natural Language Processing
(NLP) and Extractive Text Summarization algorithms to generate summarized text. Using Latent Dirichlet Allocation, a
suitable topic is assigned for the generated summarized text.
Keywords - Extractive text summarization, Abstractive Text Summarization, Latent Dirichlet Allocation, Natural Language
Processing.