Ransomware Classification and Detection with Machine Learning Algorithms

Aim:

         This study aims to improve the accuracy of Ransomware Classification and Detection with Machine Learning Algorithms Ransomware Classification, Feature Selection, Machine Learning, Cyber security

Abstract:

          Ransomware attacks pose a severe threat to the security of computer systems and data, demanding a robust and proactive approach for detection and classification. This research focuses on the application of machine learning algorithms for the effective identification of ransomware activities, employing advanced techniques such as label encoding, correlation analysis, and the utilization of the Extra Tree Classifier The study begins with the pre-processing of ransomware datasets through label encoding, facilitating the conversion of categorical data into numerical format, thus enhancing the compatibility with machine learning models. This step is crucial for training and evaluating algorithms on diverse ransomware samples. To gain insights into the relationships between different features and their impact on ransomware classification, correlation analysis is performed. By exploring the inter-dependencies among variables, the research aims to enhance the efficiency of feature selection and improve the overall performance of the machine learning models.

Proposed Method:

        This research proposes an enhancement by replacing z-score with Inter quartile range, its remove a more powerful noise and out of range. Feature selection to optimize the feature with Variance Threshold and correlation, its improving the classification accuracy and computational efficiency.