In the realm of modern service-oriented architecture, ensuring Quality of
Service (QoS) is of paramount importance. The ability to predict QoS values in
advance empowers users to make informed decisions. However, achieving accurate
QoS predictions in the presence of various issues and anomalies, including
outliers, data sparsity, grey-sheep instances, and cold-start scenarios,
remains a challenge. Current state-of-the-art methods often fall short when
addressing these issues simultaneously, resulting in performance degradation.
In this paper, we introduce a real-time QoS prediction framework (called ARRQP)
with a specific emphasis on improving resilience to anomalies in the data.
ARRQP utilizes the power of graph convolution techniques to capture intricate
relationships and dependencies among users and services, even when the data is
limited or sparse. ARRQP integrates both contextual information and
collaborative insights, enabling a comprehensive understanding of user-service
interactions. By utilizing robust loss functions, ARRQP effectively reduces the
impact of outliers during the model training. Additionally, we introduce a
sparsity-resilient grey-sheep detection method, which is subsequently treated
separately for QoS prediction. Furthermore, we address the cold-start problem
by emphasizing contextual features over collaborative features. Experimental
results on the benchmark WS-DREAM dataset demonstrate the framework’s
effectiveness in achieving accurate and timely QoS predictions.