Computers in Digital Forensics Using Machine Learning and Big Data

Abstract

The current growth of large data led to the development of AI and machine learning. The idea of strengthening the accuracy and usefulness of AI applications is also gaining popularity as big data and machine learning take off. In the realm of traffic applications, machine learning techniques improve guard safety in risky traffic situations. For vulnerable road users (VRUs), data privacy is the biggest problem with the current architectural designs. The primary cause of pedestrian traffic control failure is improper user privacy handling. The user data are vulnerable to several privacy and security flaws and are therefore at danger. If an intruder is able to break into the system, exposed data may be maliciously manipulated, manufactured, and misrepresented for illicit purposes. In this paper, a machine learning-based architecture is suggested for effectively analyzing and processing massive data in a secure setting. The suggested model takes user privacy into account when processing massive data. To achieve secure big data analytics, the suggested architecture is a layered framework with a parallel and distributed module that uses big data. The suggested architecture uses a machine learning classifier to create a unique unit for privacy management. The architecture also has a stream processing unit to process the data. Real-time datasets from a variety of sources are used to understand the proposed system, and experimental testing using credible datasets demonstrates the usefulness of the suggested architecture. Along with the training and validation outcomes, the results of the data import are also presented.