Intelligent Approaches for IoT: Water Quality Prediction

Abstract

The amount of data generated by the Internet of Things(IoT) is very large. Managing and analyzing all this data is a major challenge. Artificial Intelligence (AI) can do it faster and with greater precision. Thus, AI, and particularly Machine Learning(ML), is an effective ally for processing a growing volume of data. The objective of this work is to propose an intelligent approach for the IoT, in this case we have chosen to work on one of the applications where we can use the IoT, which is water quality prediction . We used three supervised learning algorithms K-Nearest Neighbor (KNN), Decision Tree (DT), and Random Forest (RF) on a database to develop such an approach. The RF algorithm was more efficient than KNN, and DT, as we got the highest accuracy (90%) with the RF algorithm.