Creating new services by integrating old ones is known as web service composition, or WSC. It is one of the most difficult tasks in a dynamic and distributed system. The composition process often consists of finding pre-existing services in a certain domain, selecting a suitable service, setting up the composition flow, and turning on services. The necessity of using artificial intelligence in web service composition has been the subject of extensive research in recent years. This results in a variety of solutions and creative approaches to solving this issue. The goal of our research is to use artificial intelligence to customize the makeup of web services. In our case, we used natural language processing (NLP) to examine the attributes and descriptions of existing web services. Based on the user profile, AI algorithms can select and recommend services that are appropriate for the user and meet their needs. The experimental study makes it clear that the proposed model can yield accurate results. In fact, our method achieves 99.45% accuracy and 99.74.