Collaborative Filtering Techniques for Predicting Web Service QoS Values in Static and Dynamic Environments: A Systematic and Thorough Analysis

Abstract

In recent years, the rapid growth of Web Services (WSs) has led to a proliferation of functionally similar options, making Quality of Service (QoS) a crucial factor for users in selecting the most suitable services. Predicting QoS values and recommending optimal services remain challenging, particularly in dynamic environments. This study systematically reviews QoS prediction for web services, focusing on Collaborative Filtering (CF) techniques. Following PRISMA guidelines, 512 studies were initially identified from databases like IEEE Xplore, ACM Digital Library, and Google Scholar, using keywords such as “collaborative filtering,” “web services,” and “QoS prediction.” After rigorous screening, 146 studies underwent a full-text review. Key insights were gathered on algorithms, evaluation metrics, datasets, and performance outcomes, with a focus on CF methods and advancements in hybrid and context-aware models. Despite progress, challenges in dynamic WS environments persist, highlighting the need for adaptive and real-time prediction approaches.