Résumé:
The increasing volume of textual information generated across various fields and domains
poses a significant challenge in effectively handling and extracting valuable insights from this
vast amount of data. Manual processing and analysis of every document can be time-consuming
. As a result, there is a growing need for automated systems that can provide concise summaries
of text documents, enabling users to quickly grasp the key points without having to go through
the entire content.
To address this challenge, we have proposed a text summarization system. The goal of this sys-
tem is to automatically generate condensed summaries from lengthy text documents, thereby
saving time and effort in information processing. The generated summaries should capture
the essence of the original text, providing a concise representation of the main ideas and im-
portant details. To achieve effective text summarization, the system utilizes a combination of
approaches. One of the key approaches employed is TF-IDF combined with cosine similarity.
We demonstrate the efficacy of our proposed method in summarizing articles by achieving an
impressive 70 % performance in terms of ROUGE scores. This accomplishment is based on the
evaluation of a substantial dataset consisting of 30 articles, each containing multiple pages of
content.
