Achieving Energy Efficiency in Smart Home by Using Machine Learning Algorithm for Prediction
Milka Kutseva · 2023
This article presents a system that can be utilized in Smart Homes without reducing the level of comfort for the residents. We provide a frequent sequential pattern mining approach that is appropriate for providing real-world event data for Smart Homes. With the aid of a network connection, the Smart Home allows the remote monitoring and control of equipment. In this paper we present the seven-layered IoT architecture, with the help of which the Smart Home was implemented. The main goal is to build a system that analyzes the monthly consumption data and predicts the future monthly consumption based on it. This was achieved by using machine learning approach in particular ML.NET. There are different types of machine learning algorithms that are available for this purpose. This paper focuses mainly on forecasting algorithm called Singular Spectrum Analysis (SSA).