Administrative Management Data Anomaly Access Detection Method, Based on 6G IoT

The confidentiality and safety of managerial information are now important in the rapidly changing field of digital administration. Using the improved features of 6G Internet of Things (IoT) technology, this study presents a novel approach for anomaly and illegal access detection designed for administrative management systems. The widespread use of Internet of Things (IoT) gadgets in administrative activities has made it difficult for standard security procedures to identify complex anomalies and intrusions adequately. By utilising the high-speed, quick, and enormous connection properties of 6G networks, our research seeks to close this distance. In the existing work, the CNN and RNN method is challenging for IoT devices, it works on limited resources, it is time-consuming during the data labelling. The proposed anomaly detection framework uses Autoencoders (AE) with a Multi-Layer Perceptron (MLP) method for administrative management data anomaly access detection. Initially, the dataset is collected with the help of 6G IoT devices and then pre-processed. Next the feature is compressed and extracted using Autoencoders, it compresses the input data, and it emphasises the intrinsic patterns and features to define the expected behaviour of data. Finally, the MLP is used to classify the input and normal and anomalous data. This work uses Three different datasets: NSL_KDD, UNSW_NB2015 and CICDDOD2019. The experimental results use parameters such as accuracy, f1-score, false negative rate and confusion matrix. As a result of administrative management data anomaly detection, the proposed work AE-MLP identifies breaches and classifies the normal and anomalous data efficiently.