Time series analysis and anomaly detection for trustworthy smart homes. (September 2022)

• Record Type: Journal Article
• Authors: Priyadarshini, Ishaani; Alkhayyat, Ahmed; Gehlot, Anita; Kumar, Raghvendra
• Is Part Of: Computers & electrical engineering. Volume 102(2022)
• Publication Date: 2022-09
• Subjects: Smart home -- Trustworthy systems -- ARIMA -- SARIMA -- LSTM -- Prophet; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108193
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: To establish trust in the smart home IoT environment using machine learning methods. We rely on the Change Finder algorithm to observe change points in our time series plots. The algorithm deploys a log-likelihood function on Sequentially Discounting Autoregressive (SDAR) algorithm for evaluating scores. The study incorporates several machine learning algorithms such as ARIMA, SARIMA, LSTM, Prophet, Light GBM, and VAR for the analysis. Performed and the analysis has been supported and validated using data visualization techniques as well as evaluation parameters like MSE, RMSE, and MAE. Abstract: The IoT network is expected to harbor several zettabytes of information in the future. Since trust and integrity are critical to IoT, it is essential to imbibe trust into the IoT environment for ensuring dependability and reliability. We propose a machine learning-based trustworthy system for the IoT-based smart home environment. Multiple appliances connected through the internet are susceptible to privacy issues, hence utmost care must be taken to ensure trust in the network. We consider the energy data and weather information with respect to smart homes, for comprehending the relationship between energy consumption by appliances and time period for detecting anomalous usage of appliances using the SDAR-based Change Finder algorithm. Time series analysis is performed using ARIMA, SARIMA, LSTM, Prophet, Light GBM, and VAR. The evaluation has been performed using RMSE, MSE, and MAE, and the study establishes that the ARIMA model outperforms the other models. Graphical abstract: Image, graphical abstract