Making estimations based on machine learning algorithms using data in the internet of things environment and providing complex event processing on real time data

Due to intelligent systems in the field of health, the diagnosis of diseases are made in a way that fast, is almost correct and without requiring knowledge and experience of health personnel. In order to be able to implement such a system, machine learning algorithms, which are the subbranch of data mining, are used. In this study, an intelligent system is developed by combining WEKA, which is used for machine learning algorithms, and Siddhi, which is used for real-time data processing. Internet of Things data from the field of health which can be used in the diagnosis of hernia and spondylolisthesis, were selected as the study data. The system compares Support Vector Machines, J48 Decision Tree, Artificial Neural Networks, Naive Bayes, Random Forest, jRip, kStar and Simple Logistic machine learning algorithms in terms of success and performance, using the WEKA library. At the end of this study, it is observed that the Simple Logistic algorithm gives better results than the other algorithms when the accuracy, precision, recall and fmeasure performance metrics are examined.