Internet of things (IoT) has emerged as the new trends in the wireless technology in last few years. This new area has greatly influenced the working of humans. IoT has applications in every domains of human being. The growth rate of IoT devices is exponential due to its wide applicability. Therefore, the data generated by these devices is huge and contains high variability. Such a huge amount of data needs to be modeled precisely. The effectiveness of the IoT applications lies in the preciseness of the data represented by the models. Predictive analysis helps business analysts to build models to predict trends, make tradeoff decisions, and model the real world for decision making support system. This paper presents the study on various models used for IoT data analytics. Various predictive models such as Multiple Linear Regression (LR), Support Vector Machine for regression (SVR), Random Forest (RF), Gradient Boosting Machine (GBM) and extreme Gradient Boosting Machine (XGBM) are applied on the sensor data collected from Smart Home. The comparative results produced by these models have been analyzed with reference to energy consumption and prediction. The implementation of the models is carried out on R language. The results show that XGBM model perform better based on RMSE, R-squared and MAE for given data set. It has less RMSE and high R-squared which indicates it has captured high variability in the data.
Shirke et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: