Malware Frequency Database
In computer forensics, similar to traditional forensics, hypotheses are developed based on the available facts. Despite the fact that digital evidence has been statutory witnesses for several years, there is still an argument that conclusions derived from the digital evidence that is revealed are subjective and have no scientific reasons. An increasing number of scholars have been arguing that computer forensics is computer professionals’ subjective conclusion, particularly when malware is held responsible for a felony. In this project, a system is proposed to gather data using collection mechanisms and machine learning models for predicting the malware infection rates of computers based on its features. Thus, the Neural network and Bayesian Network models have been proposed. Dataset provided from Kaggle Microsoft was used to test the accuracy of the model to predict the infection based on the feature. After conducting the experiment using the Neural network model, it was observed that the neural network has a high accuracy rate which will ensure the reliability of the result when the model feeds the Bayesian network. Therefore, the project can depend on these models for producing highly reliable probability of infection as evidential statistical that can be used in court to defeat any false claims.