Association of IoT Devices Using Fuzzy C-Means Clustering and Apriori Algorithms

Abstract

The IoT systems enable us to create various new services and offer new solutions. They play an important role in 5G and 6G systems. In order to operate IoT systems efficiently and support scalability, they should be grouped properly. There are many clustering and association ways for IoT systems. Clustering algorithms of machine learning are techniques for finding similarity groups in data. In this paper, we propose two-steps IoT device association method by combining Fuzzy C-Means clustering algorithm and Apriori algorithm. Fuzzy C-Means clustering as a soft clustering approach assigns data points to probability score and allocates them to clusters. Apriori algorithm is used in finding frequent sets and relevant association rules. As combining the clustering technique and the association rule mining technique, we can obtain multiple benefits including handover cost reduction. The performances of the proposed method are evaluated and compared with K mean clustering algorithm in terms of energy efficiency and handover frequency.