Service-based systems (SBSs) dynamically integrate various third-party services at runtime, offering sophisticated and adaptable functionalities. However, this dynamic integration introduces unpredictability and uncertainty, posing challenges for system maintenance. Proactive self-adaptation based on prediction has emerged as a common approach to address these issues. Many existing methods focus primarily on prediction accuracy, often neglecting the crucial aspect of “earliness“ in predictions. Balancing earliness and accuracy is very crucial in practice, as it provides more time for self-adaptation while ensuring reliable predictions. Therefore, we propose PSA4SBS (Proactive Self-Adaptation for Service-Based Systems), which aims to balance this trade-off in SBSs. PSA4SBS features an ensemble prediction model based on deep neural networks (DNN) and an improved long short-term memory (ILSTM) architecture. It helps SBSs adapt to the frequent unpredictability and instability, facilitating the achievement of adaptation objectives. This mechanism im-proves the prediction and analysis of adaptation goal violations, thereby enhancing the reliability and performance of service level agreements (SLAs) for quality of service (QoS). We evaluated PSA4SBS's performance in a decentralized tele-assistance system using four key metrics. Preliminary experimental results show its exceptional performance.

Abstract Bug triaging is a vital process in software maintenance, involving assigning bug reports to developers in the issue tracking system. Current studies predominantly treat automatic bug triaging as a classification task, categorizing bug reports using developers as labels. However, this approach deviates from the essence of triaging, which is establishing bug–developer correlations. These correlations should be explicitly leveraged, offering a more comprehensive and promising paradigm. Our bug triaging model utilizes graph collaborative filtering (GCF), a method known for handling correlations. However, GCF encounters two challenges in bug triaging: data sparsity in bug fixing records and semantic deficiency in exploiting input data. To address them, we propose PCG, an innovative framework that integrates prototype augmentation and contrastive learning with GCF. With bug triaging modeled as predicting links on the bipartite graph of bug–developer correlations, we introduce prototype clustering‐based augmentation to mitigate data sparsity and devise a semantic contrastive learning task to overcome semantic deficiency. Extensive experiments against competitive baselines validate the superiority of PCG. This work may open new avenues for investigating correlations in bug triaging and related scenarios.