Missing IoT Data Prediction with Machine Learning Techniques

Every day, the amount of data generated by industrial applications based on Internet Things (IoT) grows. However, acquired as a result failures and communication disconnections in IoT devices might be noisy, inaccurate, incomplete. These issues have become crucial for production, quality, processing, analysis. The datasets used scope this study were collected real-time from water neutralizer system Sivas Numune Hospital, which converts medical waste into household waste. Medical liquid wastes hospitals are exposed to chemical neutralization process means pH change with before being transferred sewer. In regard, monitoring levels is environmental protection. aspect, two varying quantities missing evaluated prediction PH using linear regression (LR), support vector machines (SVM), k-nearest neighbor (KNN), random forest (RF), decision tree (DT) machine learning algorithms. Mean absolute error (MAE), mean squared (MSE), root square (RMSE) performance metrics evaluate As consequence analysis, it was determined that SVM algorithm performed better distinct datasets. evaluation indicates algorithms remarkably efficient at predicting data.