My project focuses on ransomware detection using machine learning and Python. The primary objective is to identify ransomware present in specific systems by leveraging advanced machine learning techniques. Unlike traditional methods, this project incorporates Natural Language Processing (NLP) to enhance the accuracy and efficiency of the detection model. By using NLP techniques such as normalization, vectorization, and tokenization, the project aims to improve the training and testing phases of the model, resulting in better detection capabilities compared to pre-built models. The achieved accuracy of the model stands at 78%.
I decided on making this project so that the detection of ransomware is local to the individual system and hence not sharing any important information to any internet security platforms. Its really helpful in crowded places like library, schools, colleges, etc to use my project so that easily detection of ransomware or malware is done before it attacks many other systems connected to the same networks. The model actually understands how often vulnerabilities occur and specifically at what time, so that the software becomes alert and notifies the user about the vulnerability.
The ransomware detection system starts by collecting data that may contain potential ransomware samples. This data undergoes preprocessing through NLP techniques. Normalization standardizes the data format, making it easier to analyze. Tokenization breaks down the data into manageable pieces, and vectorization converts these tokens into numerical vectors suitable for machine learning algorithms. These processed vectors are then used to train a machine learning model, which learns to distinguish between normal and ransomware-infected data. During testing, the trained model evaluates new data samples to detect any signs of ransomware. The use of NLP significantly enhances the model's ability to accurately identify ransomware, achieving a commendable accuracy rate of 78%.
https://drive.google.com/file/d/1lMh9CQ9fVz2-CmrfABeo_lSt8eMrZnyx/view?usp=sharing (Link for project execution)
Comments