eversvd,+3 (18)

Since the dawn of the Internet, we have been inundated with an excess of information. The volume of information available on the Internet is expected to grow exponentially, necessitating effective summarization techniques. This study aims to address the issues of information overload and media credibility by condensing news articles related to the Tamilnadu Legislative Assembly election. Data was gathered over two months from various news sources in Tamil, which were later translated to English. K-means clustering was employed to segregate Tamilnadu-related articles, followed by a pre-processing phase to eliminate translation errors and ambiguities. Individual articles were summarized using a linear regression model focusing on features such as named entities and word similarity. Finally, the summaries were further condensed using BERT extractive summarization to reduce redundancy. Evaluation revealed a precision score of 0.512, a recall of 0.25, and an f-measure of 0.31 when comparing generated summaries with corresponding articles in cases where the introduction was absent. This research highlights the potential of summarization methodologies in enhancing the efficiency of information retrieval in the context of news articles.