TY - JOUR T1 - Artificial Intelligence Helps Protect Smart Homes against Thieves TT - Artificial Intelligence Helps Protect Smart Homes against Thieves AU - Pala, Zeydin AU - Özkan, Orhan PY - 2020 DA - September DO - 10.24012/dumf.700311 JF - Dicle Üniversitesi Mühendislik Fakültesi Mühendislik Dergisi JO - DUJE PB - Dicle University WT - DergiPark SN - 1309-8640 SP - 945 EP - 952 VL - 11 IS - 3 LA - en AB - Interaction with the environments in which humans live is increasing more and more, and Artificial Intelligence (AI) offers significant contributions to this. Although the topic of smart homes has attracted a great deal of attention from researchers, the AI-based application in this area is still in its infancy. In this study, a home security automation system, which is quite simple, but smart and AI-based, is proposed. When the home-dwellers were not at home, the home lighting system tried to be managed with AI at night, as if life was still there. The AI-based smart home physical design was done using Arduino equipment and was tried to be adapted to the real-life environment with software support. As if there was someone at home, a special dataset, which was consisted of nine inputs, one output vector and about 5500 samples was created to turn on/off the home lights in a manner suitable for night life. The home lighting system was successfully managed using an AI-based system that learns nightlife lighting habits. The proposed system performance was tested in support of commonly used machine learning classification algorithms such as Multi-layer perceptron (MLP), Linear support vector machine (L-SVM), Gaussian Naive Bayes (NB), and linear discriminant analysis (LDA). The accuracy values of MLP, L-SVM and NB algorithms were 96.69%, 94.98% and 91.23%, respectively. Our results show that a home with AI could be safer and more secure against theft. KW - Artificial intelligence KW - machine laerning KW - classification KW - smart home KW - security KW - theft N2 - Interaction with the environments in which humans live is increasing more and more, and Artificial Intelligence (AI) offers significant contributions to this. Although the topic of smart homes has attracted a great deal of attention from researchers, the AI-based application in this area is still in its infancy. In this study, a home security automation system, which is quite simple, but smart and AI-based, is proposed. When the home-dwellers were not at home, the home lighting system tried to be managed with AI at night, as if life was still there. The AI-based smart home physical design was done using Arduino equipment and was tried to be adapted to the real-life environment with software support. As if there was someone at home, a special dataset, which was consisted of nine inputs, one output vector and about 5500 samples was created to turn on/off the home lights in a manner suitable for night life. The home lighting system was successfully managed using an AI-based system that learns nightlife lighting habits. The proposed system performance was tested in support of commonly used machine learning classification algorithms such as Multi-layer perceptron (MLP), Linear support vector machine (L-SVM), Gaussian Naive Bayes (NB), and linear discriminant analysis (LDA). The accuracy values of MLP, L-SVM and NB algorithms were 96.69%, 94.98% and 91.23%, respectively. Our results show that a home with AI could be safer and more secure against theft. CR - 1. Benjamin K., Sovacool, Dylan D., Furszyfer DR. Smart home technologies in Europe: A critical review of concepts, benefits, risks and policies, Renewable and Sustainable Energy Reviews, Volume 120, 2020, pp.1-20. CR - 2. Skn, H., Kalkan, H., Cetili, B. Classification of physical activities using accelerometer signals. In: Signal Processing and Communications Applications Conference, Mugla,Turkey,2012, pp.1-4. CR - 3. Gne, H., Orta, E., Akda, D. Developing synthetic data generation software for artificial intelligence techniques used in smart home systems. Journal of Balikesir University Institute of Science and Technology, 18(2):1-11,2016. CR - 4. Gariba, D., Pipaliya, B. Modelling human behaviour in smart home energy management systems via machine learning techniques. In: Proceddings of 2016 International automatic control conference,Taichung, Taiwan, 2016, pp. 53-58. CR - 5. Dixit, A., Naik, A. Use of prediction algorithms in smart homes. International Journal of Machine Learning and Computing, 4:157-162, 2014. CR - 6. Collado-Villaverde, A., R-Moreno, MD., Barrero, DF., Rodriguez, D. Machine learning approach to detect falls on elderly people using sound. In: Benferhat S., Tabia K., Ali M. (eds) Advances in Artificial Intelligence: From Theory to Practice. IEA/AIE. Lecture Notes in Computer Science, Springer, Cham, 2017, pp. 149-158. CR - 7. Alshammari, T., Alshammari, N., Sedky, M., Howard, C. Evaluating machine learning techniques for activity classification in smart home environments. World Academy of Science, Engineering and Technology International Journal of Information and Communication Engineering, 12 (2):58-64, 2018. CR - 8. Alhafidh, BAH., Allen, WH. Comparison and Performance Analysis of Machine Learning Algorithms for the Prediction of Human Actions in a Smart Home Environment. In:Proceedings of the International Conference on Compute and Data Analysis, New York, USA, 2017, pp. 54-59. CR - 9. Dahmen, J., B. Thomas, DJ. Cook,X.Wang, Activity Learning as a Foundation for Security Monitoring in Smart Homes. Sensors ,17(4):pp. 1-17, 2017. CR - 10. Cumin, J., Lefebvre, G., Ramparany, F., Crowley, FL. Human activity recognition using place-based decision fusion in smart homes. In:International and Interdisciplinary Conference on Modeling and Using Context, Paris, France, 2017, pp.137-150. CR - 11. Singh, D., Merdiva, E., Hanke, S., Kropf, J., Geist, M., Holzinger, A. Convolutional and recurrent neural networks for activity recognition in smart environment. In:Towards integrative machine learning and knowledge extraction, Banff, AB, Canada, 2015, pp.194-205. CR - 12. Jebakumari, VS., Shanthi, D., Sridevi, S., Meha, P. Performance evaluation of various classification algorithms for the diagnosis of Parkinson’s disease. In:Intelligent Techniques in Control, Optimization and Signal Processing (INCOS), Srivilliputhur, India, 2017, pp.1-7. CR - 13. Cover, TM., Hart, PE. Nearest neighbor pattern classification. IEEE Transactions on Information Theory, IT13(1):2127, 1967. CR - 14. Bhatia, N. Survey of nearest neighbor techniques. International Journal of Computer Science and Information Security, 8(2):302-305, 2010. CR - 15. Mohana, TK., Lalitha, V., Kusuma, L., Rahul, N., Mohan, M. Various Distance Metric Methods for Query Based Image Retrieva. International Journal of Engineering Science and Computing, 7(3):5818-5821, 2017. CR - 16. Zhou, ZH. Ensemble methods foundations and algorithms, Boca Raton, FL,2012. CR - 17. Marshald, S. Machine learning an algorithmic perspectives, Boca Raton, FL, 2015. CR - 18. Rogers, S., Girolami, M. A first course in machine learnings, Boca Raton, FL, 2012. CR - 19. Sarkar, D., Bali, R., Sharma, T. Practical Machine Learning with Python, Bangalore, Karnataka, India, 2018. CR - 20. Vijayarani, S., Dhayanand, S. Data mining classification algorithms for kidney disease prediction, International Journal on Cybernetics & Informatics (IJCI), 4(4):13-25, 2015. CR - 21. Mazzolenia, M., Previdia, F., Bonfiglio, NS. Classification algorithms analysis for brain computer interface in drug craving therapy. Biomedical Signal Processing and Control 2017, https://doi.org/10.1016/j.bspc.2017.01.011. CR - 22. Takci, H. Improvement of heart attack prediction by the feature selection methods, Turk J Elec Eng & Comp Sci,26:1-10, 2018. UR - https://doi.org/10.24012/dumf.700311 L1 - https://dergipark.org.tr/en/download/article-file/997310 ER -