The development of search engines that can retrieve relevant content based on user-specified keywords is necessary due to the increasing volume of digital information available on the internet. In this abstract, a novel approach is presented to build a keyword-based search engine that uses cosine similarity to enhance search result relevance and precision. Our search engine, which is keyword-based and uses cosine similarity, has several benefits over traditional search engines. The accuracy of search results is improved by taking into account semantic relationships between words, leading to a more contextually relevant ranking. With increased precision and efficiency, users can expect better search experiences when searching for the information they need.