A Frequency-Driven Approach for Extractive Text Summarization

Due to Digital Revolution, the majority of books and newspaper articles are now available online. Particularly for kids and students, prolonged screen time might be bad for eyesight and attention span. As a result, summarizing algorithms are required to provide long web content in an easily digestible style. The proposed methodology is using term frequency and inverse document frequency driven model, in which the document summary is generated based of each word in a corpus. According to the preferred method, each sentence is rated according to its tf-idf score, and the document summary is produced in a fixed ratio to the original text. Expert summaries from a data-set are used for measuring precision and recall using the proposed approach's ROUGE model. towards development of such a framework is presented.