Machine Learning Algorithms to Evaluate Fuzzy Logic Web Services for Monitoring the Real-Time Applications

Most people, particularly those above the age of 65, are vulnerable to serious injury as a result of fall. Usually, falls go unreported and are dismissed as an unavoidable occurrence. In elderly adults, a variety of wearable fall detection systems have now been developed. The majority of these gadgets, however, rely on regional information processors. Wearable technologies employ a variety of methods to detect fall on real time; it emphases fall detection vfuzzification, Java Fuzzy Modeling Language (JFML), and service-oriented architecture, and is based on IEEE 1855-2016. Furthermore, to avoid such mishaps, multiple methodologies were employed that identify a fall employing machine learning algorithms and human movement location information. Wearable acceleration & gyroscope detectors are used to evaluate fuzzy logic web services, to distinguish among the fall & a non-fall, machines learning methods including such k-NN, decision tree, randomized forests, & high gradients boosting are employed. The study will compare real-time practical fall monitoring systems with fuzzy services and machine learning to see which is well suited for real-time fall detection. The suggested fuzzy logic web services will efficiently discriminate between fall and non-fall events in a real-world context, according to study results, with precision, sensitivities, and specificities of 90 percent, 88.89 percent, and 91.67 percent, respectively. Using Machine learning’s randomized forests method scored 99.19 percent, 98.3 percent, and 99.3 percent, respectively.