Core Concepts
The UGRansome2024 dataset, optimized through feature engineering, enables highly accurate ransomware detection using the Random Forest algorithm, with insights into the financial impact of different ransomware variants.
Abstract
The study introduces the UGRansome2024 dataset, an optimized version of the original UGRansome dataset, for ransomware detection and classification. The dataset was refined through feature engineering techniques to enhance its effectiveness in capturing relevant patterns and characteristics of ransomware behavior.
The researchers applied the Random Forest algorithm to the UGRansome2024 dataset, achieving a classification accuracy of 96%. The analysis revealed that certain ransomware variants, such as those utilizing Encrypt Decrypt Algorithms (EDA) and Globe ransomware, had the highest financial impact, quantified in bitcoin (BTC) amounts.
The findings highlight the importance of machine learning in ransomware detection and mitigation, emphasizing the value of comprehensive datasets like UGRansome2024 in developing robust cybersecurity strategies. The study also discusses the limitations of the current approach and provides recommendations for future research, including expanding the dataset, exploring alternative detection methods, and addressing the evolving nature of ransomware threats.
Stats
Encrypt Decrypt Algorithms (EDA) and Globe ransomware incidents resulted in the highest financial impact, quantified in bitcoin (BTC) amounts.
Quotes
"Achieving a classification accuracy of 96% underscores the efficacy of machine learning techniques in identifying ransomware attacks, particularly those involving Unusual Transaction patterns."
"The identification of EDA and Globe ransomware as having the highest financial impact highlights the importance of understanding specific ransomware variants and their associated risks."