IoT Content Cache Efficacy in Health Care Data Diagnostics Using Isotonic Regressive Adaptive Boost Classification

Abstract

Information-centric networking (ICN) with the internet of health things (IoHT) provides a secure network between medical experts and patients. Several researchers have investigated cache with IoHT data in ICN. However, there are still issues like which content to be cached or node selection for placing the content. In this work, an efficient caching strategy called, isotonic regressive adaptive boost classification-based information centric network (IRABC-ICN) is proposed with IoHT data. Initially, patient healthcare data requests are obtained from maternal healthcare. The patient information is registered and an information copy is deployed to the cache of every router node. Each request to the router node is searched in the content storage (CS). Next, isotonic regressive analysis is to analyze patient healthcare data request search. Here CS is checked and upon successful validation, the particular router node is selected to place the content and on contrary, the patient information is stored in the cache. Finally, the modest adaptive boost classification is carried out to obtain strong classification results with higher efficiency of content distribution and lesser network traffic, and server bandwidth. To prove the performance of this proposed IRABC-ICN, its cache hit rate, network latency, and average request length are used as the evaluation metrics. The efficacy of IRABC-ICN is proved by comparing it with existing PCSRC, cooperative caching, and context-based caching mechanisms with the maternal health risk dataset. The outcome of the proposed IRABC-ICN achieves enhancement in cache hit rate by 9%, minimization of network latency, and average request length by 38%, and 13% as compared to PCSRC, cooperative caching, and context-based caching mechanism respectively.