Determining Malware Attacks in Iot Health Care System Using Self Attention Based Depth Poly O (Optimized)-network

Abstract The Internet of Things (IoT) is becoming more important in numerous sectors, including healthcare, industry, the military, and education. The framework successfully safeguards the privacy, authenticity, and accessibility of data in a networked setting. Health care is only one of many sectors that may benefit from the given solutions, which are essential for protecting the personal information of patients and maintaining the integrity of their medical records. The privacy, security, and reliability of the whole health care system online are still at danger owing to a broad variety of intermediary assaults and infiltration activities, despite the fact that the Internet of Things (IoT) offers trustworthy mechanisms for keeping data secure. In this research, we employ a hyper-tuned optimal classifier based on deep learning to overcome these issues. Here the real time patients sensor data are retrieved and it can be preprocessed for removing the error by using the Minmax Hat normalization. Then the features can be selected using wrapper discriminant component analysis. Then the proposed classifier can be optimized using the Hybrid CUADA (Cuckoo Adam) optimization algorithm. After classification parameter optimization, the collected features are fed into the newly-introduced self-attention based depth poly O (optimized)-Network to detect malware intrusions and monitor patient records. The system's effectiveness has been assessed based on experimental findings and subsequent discussions.