A Lightweight Association Rules Based Prediction Algorithm (LWRCCAR ) for Context-Aware Systems in IoT Ubiquitous, Fog, and Edge Computing Environment

Abstract- Proactive is one main aspect of ubiquitous context-aware systems in IoT environment. The power of artificial intelligent is employed to realize this high-end aspect. Ubiquitous context-aware systems in IoT environment needs a light-weight intelligent prediction techniques especially within fog and edge computing environment where technologies capabilities are poor. On the other hand, in big data area the amount of data used to train ubiquitous context-aware systems is huge. This paper suggests a light-weight prediction algorithm to help such systems to work more effectively. The proposed algorithm is an improvement of the RCCAR algorithm that we developed in previous work. RCCAR utilizes association rules for prediction. The contribution of this paper is to minimize the number of association rules by giving a priority to associations that produced of high order. The prediction is scored and formulated mathematically using the confidence measure of association rules. A real-world dataset is used to evaluate the proposed algorithm in various scenarios. The results show that the proposed algorithm achieves better prediction scores. For future work, extensive experiments with many datasets is recommended.