The summer heat of cryptojacking season : Detecting cryptojacking using heatmap and fuzzy

• Main Authors: Ahmad Firdaus, Zainal Abidin, Aldharhani, Ghassan Saleh, Zahian, Ismail, Mohd Faizal, Ab Razak
• Format: Conference or Workshop Item
• Language: English; English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Subjects: QA75 Electronic computers. Computer science; QA76 Computer software; T Technology (General); TA Engineering (General). Civil engineering (General)
• Online Access: http://umpir.ump.edu.my/id/eprint/39105/1/The%20summer%20heat%20of%20cryptojacking%20season_Detecting%20cryptojacking.pdf; http://umpir.ump.edu.my/id/eprint/39105/2/The%20summer%20heat%20of%20cryptojacking%20season_Detecting%20cryptojacking%20using%20heatmap%20and%20fuzzy_ABS.pdf; http://umpir.ump.edu.my/id/eprint/39105/; https://doi.org/10.1109/ICCR56254.2022.9995891

Cryptojacking is a subset of cybercrime in which hackers use unauthorised devices (computers, smartphones, tablets, and even servers) to mine cryptocurrencies. Similar to many other forms of cybercrime, the objective of cryptojacking is achieve profit illegally. It is also designed to remain entirely concealed from the victim's view. However, its attacks continue to evolve and spread, and their number continues to rise. Therefore, it is essential to detect cryptojacking malware, as it poses a significant risk to users. However, in machine learning intelligence detection, an excessive number of insignificant features will diminish the detection's accuracy. For machine learning-based detection, it's important to find important features in a minimal amount of data. This study therefore proposes the Pearson correlation coefficient (PMCC), a measure of the linear relationship between all features. After that, this study employs the heatmap method to visualise the PMCC value as a colour version of heat. We utilised The Fuzzy Lattice Reasoning (FLR) classifier for classification algorithms in machine learning. This experiment utilised actual cryptojacking samples and achieved a 100 percent detection accuracy rate in simulation.