An Efficient Anomaly Detection Through Optimized Navigation Using Dlvq-Cdma And H-Dso In Healthcare Iot Environment

Abstract

An Anomaly detection (AD) framework intends to discover irregular data and also unusable activities in a system. The abnormality in the healthcare information is picked up by the AD in the healthcare system and then, the outcome is updated for the authority to evaluate the data. Numerous researchers have developed an AD method that has the disadvantage of data loss issues and complexity in computation. An enhanced AD framework utilizing Deep Learning Vector Quantization-Correlation Distance Mayfly Algorithm (DLVQ-CDMA) and Hyper-sphere Dolphin Swarm Optimization (H-DSO) methodology is presented in this work to overcome these disadvantages. By aid of the Internet of Things (IoT)-connected systems, proffered model gathers information about the patient and as well forwards the information to patient's health care application. Information from health care application is then sent via the optimal path by utilizing the H-DSO method. The data is uploaded to the cloud server later and then, it is recovered and provided to the AD system. The data is then pre-processed in an AD system. After extricating the features, the feature reduction is performed by employing the Entropy-Generalized Discriminant Analysis(E-GDA) scheme. Subsequently, the DLVQ-CDMA algorithm is utilized with the required features. Information is formerly categorized as usual data or irregularity data. data, which is attacked is stored in the log file and the normal data will undergo further evaluation for the identification of the presence of disease or disorder. After evaluation, the outcome is communicated to the patient. The experiential analysis specifies that the proffered DLVQ-CDMA methodology executes better than the prevailing methodologies.