Context-Aware Machine Learning: A Survey

This survey investigates advancements in Context-Aware Machine Learning, emphasising the critical role of contextual data in achieving accurate interpretation and decision-making. It highlights the diverse nature of contextual information and their relevance across spatial, temporal, and spatio-temporal domains. Through a systematic literature review, the role of context mining, modeling, and management for the definition of formal contextual rules was discovered. Moreover, the study evaluates various context integration methods and context-aware learning architectures, including modular, hybrid, ensemble, and graph-based learning. Ensemble learning is identified as particularly effective for integrating diverse contextual data, thereby enabling the use of future technologies, such as the integration of quantum artificial intelligence.