Customer Identification in Healthcare using an IoT-based Multimedia Traffic Categorization Method

Parkinson's disease is a neurological ailment that impairs mobility in humans that is caused by a tolerant dysfunction of the nervous system. It progresses slowly, with the onset of a barely noticeable tremor in only one hand every now and again. Nevertheless, while tremors are perhaps the most well-known symptom of PD, the ailment is also associated with stiffness and slowness of movement. Three well-established machine learning-based algorithms are used in this study to solve the challenge of enhanced categorization of data accessible at an Internet of Things node. After reviewing current literature, it has been discovered that just a few papers have been published on the categorization and forecasting of IoT-based data using ML approaches. A growing amount of attention has been paid in recent times to the categorization and forecasting of health information that is done online. In light of this aim, the current study investigates the intelligent categorization of PD using data from the Internet of Things (IoT), which is accomplished through the use of ML algorithms. The ML based classifiers employed in this article are the Decision Tree, the Random Forest, and the Naive Bayes. These classifiers were chosen based on their stable and better classification results for other common datasets, which is demonstrated in previous research. The Internet of Things-based node gets the data and provides a categorization answer more quickly, so assisting in choice. The simulation-based studies are carried out with the help of the two representative data sets that have been provided. The findings of the simulations are also used to investigate the relationship between the amount of features and classification accuracy. The results reveal that the Random Forest, Decision Tree, and Naive Bayes are the top three classification algorithms in terms of classification accuracy. Naive Bayes, Decision Tree, and Random Forest are the models that rank highest in terms of execution time, while the others are ranked lower. The right classifier for usage in Internet of Things (IoT) based industrial contexts must be determined based on the requirements.