In recent years, the growth of Internet of Things (IoT) as an emerging has been unbelievable. The number of networkenabled devices in IoT is increasing dramatically, leading to the massive production of data. These data contain valuable information which can be used in areas, such as science, industry, business and even social life. To and analyze this information and make IoT systems smart, the only is entering artificial intelligence (AI) world and leveraging the power machine learning and deep learning techniques. This paper evaluates the of 11 popular machine and deep learning algorithms for task using six IoT-related datasets. These algorithms are according to several performance evaluation metrics including, recall, f1-score, accuracy, execution time, ROC-AUC score and matrix. A specific experiment is also conducted to assess the speed of developed models. The comprehensive experiments indicated, considering all performance metrics, Random Forests performed better than machine learning models, while among deep learning models, ANN and CNN more interesting results.
No takes yet. Share an insight, caveat, or question.
Vakili et al. (2020) studied this question.