Novel Predictive Analyzer for the Intrusion Detection in Student Interactive Systems using Convolutional Neural Network algorithm over Artificial Neural Network Algorithm

The purpose of this project is to improve the detection of infiltration in interactive systems by employing convolutional neural networks (CNN) in comparison to other machine learning techniques. The Suggested Method: Using an unique Convolutional neural network, we were able to identify intrusions in the interactive system. These tests were carried out with a sample size of ten individuals for each, and the software tool jupyter notebook was utilised. Calculations are made to determine the accuracy values for the identification of incursions in order to measure the performance of the convolutional neural network in comparison to artificial neural networks. Discussion and the Results: The predictive analysis of intrusion detection in the student interactive system using ANN and how it compares to using CNN The accuracy of 93% through ANN may be obtained by the utilisation of statistical analysis, whilst the accuracy of 91% through convolution neural networks can be achieved. The findings illustrate a high degree of dependability across independent variables, with a confidence interval of 95%. When compared to artificial neural networks, the performance that convolutional neural networks provide for predicting intrusion in student interactive systems is much superior.