Study of Machine Learning Models for IoT Based Efficient Classroom Usage

This paper presents performance analysis and comparison of machine learning algorithms for future use in a smart campus framework. The following error rates, such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Square Error (MSE) and R squared error are considered for models such as Random Forest (RF), Multiple Linear Regression (MLR), Decision Tree Regression (DTR), Support Vector Regression (SVR), Polynomial Regression (PR), Generic Predictive Computation Model (GPCM). The investigation how to reduce the processing time for the algorithms is presented. The following error rates such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Square Error (MSE) are considered for Random Forest, Multiple Linear Regression, Decision Tree Regression, Support Vector Regression, Polynomial Regression models and Machine Learning tools taken from Use Cases of Generic Predictive Computation Model (GPCM) are partially applied. Testing with our arbitrary data will be conducted. A lower error rate for selected algorithms with reduced number of parameters (5 parameters) as opposed to 11 parameters is achieved.