Integration of K-Means Clustering and Naïve Bayes Classification Algorithms for Smart AC Monitoring and Control in WSAN

Abstract

There is still a lot of excessive use of lamps, televisions, air conditioners (AC) and other electronic goods, resulting in a surge in electricity bills charged by electricity users due to neglect and waste of electrical energy. Based on these problems a system that is needed not only to monitor but also to control electrical equipment remotely so that electricity consumption can be controlled. Wireless Sensor and Actuator Network (WSAN) technology can monitor the physical condition of the environment which is widely applied in intelligent environments. WSAN is placed at certain regional points that will be observed the physical condition of the environment, each WSAN can use several sensors and actuators which will later be sent to the server via a wireless connection. In this research, we will test by making Smart AC (Air Conditioner) where at every point where there is AC will be installed WSAN. Data from several sensors generated from WSAN will be sent to the server to be observed and processed using intelligent computing and machine learning (K-Means and Naïve Bayes) so that the AC can turn on and off according to the physical conditions in the place. We evaluate our model too by using Confusion Matrix and obtain the score for accuracy 90%, precision 83%, recall 100% and error rate 10%, So our model can be called good model.