Process-Specific Extensions for Enhanced Recommender Systems in Business Process Management

In Business Process Management (BPM) the integration of advanced recommender systems emerges as a critical strategy to enhance process efficiency and user satisfaction. Despite the dissemination of these systems, there remains a distinct lack of incorporating execution relevant context data, particularly those generated during process execution and prevailing environmental conditions. This paper addresses this gap by proposing process-specific extensions for augmenting an existing recommender system framework. Our approach not only enhances the adaptability and accuracy of recommendations but also sustains the applicability of existing algorithms, ensuring seamless integration into available recommender frameworks. The potential of this refined approach is demonstrated by evaluating a process scenario based on synthetic data.