Comparative Analysis of TextRank with GloVe and K-Means Clustering for Extractive Summarization

In natural language processing, text summarization is a crucial task that attempts to simplify large amounts of text while preserving important information. In order to summarize BBC news articles in the fields of business, entertainment, politics, sports, and technology categories, this research compares the TextRank and K-Means algorithms. TextRank performs well in tech and politics (0.56 and 0.52) based on BLEU scores, but K-Means does better in entertainment (0.65). They both do well in business (0.53) and sports (TextRank: 0.51, K-Means: 0.54). By using content characteristics to inform algorithm selection, the findings improve summarization and natural language processing methods.