<p>The proliferation of web services, along with their increasing adoption by prominent Internet companies and a substantial user base, has led to a heightened focus on them within the industry and the research community. This growing dependence necessitates accurate reputation assessment to ensure reliability and trustworthiness. Traditional models rely heavily on user ratings, which are often subject to bias and manipulation. In this paper, we propose a novel model for web services reputation assessment based on social network data. The proposed model integrates quantitative metrics, such as service activity and popularity, with qualitative sentiment analysis of user comments, and machine learning-based filters are employed to enhance reliability. Experimental evaluations using real-world Twitter datasets demonstrate the effectiveness of the proposed approach. We compared our approach with an LSTM-based reputation model, and the results show that our model achieves higher accuracy but has a slightly higher execution time. Future work will focus on expanding datasets, improving real-time adaptability, and incorporating additional trust metrics.</p>