This study describes a system that uses NLP-based summarizing to streamline information retrieval. In order to extract the most pertinent and instructive sentences from the input text, our suggested approach makes use of features such as cosine similarities, PageRanking duplicate word, concentrating mapping, data clearing, stop phrase removal, word count. The methodology entails preparing the input text, determining each sentence's cosine similarity to the raw text, and sorting the sentences according to their word counts. Our approach achieves great accuracy in summarizing the major ideas of the documents while keeping coherence and readability when tested on a sample of various texts and review based on customers option. Unsupervised learning techniques like Text rank are employed for extracting summarization of texts. Typically, text summaries are produced solely using the text rank algorithm. However, in this study, summaries are extracted using the text rank algorithm in conjunction with cosine similarity. Our method is able to recognize significant and succinct statements because a word count and cosine similarities scores for the top-ranked phrases are 0.46. Overall, by quickly and accurately summarizing vast amounts of text using NLP approaches, our suggested system can greatly increase the effectiveness of information retrieval tasks.