Summarization of Daily News Using TextRank and TF-IDF Algorithm

In the current era of digitization, the information gets generated day by day and WWW is overflooded with digital information. This information is used for various purposes like reading, engineering, medical, civil, research, etc. Filtering the important information from existing one is called summarization. Manual summarization is one of the tedious as well as challenging tasks as sometimes the results may be biased. It generates the need for a system that can automatically summarize the text without vanishing its meaning. In the proposed paper authors are summarizing daily news using two algorithms—TextRank and TF-IDF. Unsupervised approach is used for summarization process. Text summarization techniques are discussed. Initially both the algorithms are elaborated, and then for experimental purpose CNN-Daily News Dataset from Kaggle is supplied to both the algorithms. Finally generated summaries are evaluated based on Rouge scores.