Key points are not available for this paper at this time.
In this fast moving world, everybody has the urge to get updated to what's happening around the world. There is a heavy demand for knowing information which is crisp and to the point without getting reading the may be ten articles on it, but to get what the ten articles mean. So, to solve this issue, here we present a novel approach for 'Summarizing the Discussion' which is aimed at presenting the user with the most relevant summary containing all the important points of the discussion. At this juncture, we need to consider the polarity of comments and the whole discussion and present the user with a short, unbiased, comprehensive and non-redundant summary. Our proposal consists of a Nested Thematic Clustering approach coupled with Keyword or Key phrase extraction and sentiment analysis or rather polarity calculation. Be it political, social, geographical, philosophical, superstitious, educational, and environmental or entertainment domain, all the news is generally shared through forms or posts of internet. Now, it gets a little obvious that different people have different views. Hence, we make it a point to include and ensure the multi-dimensionality of every thread under the post. We have used an extractive method of summarization with the help of Natural Language Processing where we incorporate the techniques of repeated clustering and ranking. The proposed model is able to generate a desirably relevant summary.
Lalithamani et al. (Fri,) studied this question.