MinerGuard: A Solution to Detect Browser-Based Cryptocurrency Mining through Machine Learning

Wu, Min-Hao; Lai, Yen-Jung; Hwang, Yan-Ling; Chang, Ting-Cheng; Hsu, Fu-Hau

doi:10.3390/app12199838

Cited by 6 publications

(4 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) Rogue Web Mining Programs Hackers also attempt to make a fortune by mining cryptocurrencies. In recent years, a great number of rogue web-mining programs have emerged [57], and they have utilized users' computers to mine from popular websites, which is impossible to defend against effectively. Coinhive is an example of a rogue mining program [57].…”

Section: Common Virus Attack Methodsmentioning

confidence: 99%

“…In recent years, a great number of rogue web-mining programs have emerged [57], and they have utilized users' computers to mine from popular websites, which is impossible to defend against effectively. Coinhive is an example of a rogue mining program [57]. The programs mining Monero are packed in JavaScript at the browser end, and website dealers are able to utilize the computing power of users' computers to mine Monero by embedding Coinhive programming code in their websites [58].…”

Section: Common Virus Attack Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Identification of SMEs in the Critical Factors of an IS Backup System Using a Three-Stage Advanced Hybrid MDM–AHP Model

Chen

Chou

Lin³

et al. 2023

Sustainability

View full text Add to dashboard Cite

Backup system work represents “the last mile” of information security (IS). To avoid data loss or damage, enterprises should execute data backup periodically to ensure the integrity and availability of such data. Additionally, due to the continuous emergence of IS incidents featuring malicious attacks in recent years, major firms in countries around the world have successively reported being under attack by ransomware viruses. In particular, small and medium enterprises (SMEs) became the potential targets of malicious attacks based on their different types of IS awareness and degrees of digitalization; therefore, IS work has become one of the essential topics with special significance for numerous SMEs. To this end, this paper studied the factors influencing SMEs’ adoption of IS backup systems in the hope that the critical decision-making behaviors of SMEs regarding the issue of IS could be learned. Practical suggestions can be made for the marketing schemes adopted by IS manufacturers concerning the planning of IS backup systems. Thus, this study used three methodological stages to address the exciting issue of IS backup systems for SMEs. In the first stage, 11 factors at two hierarchies involving three constructs influencing SMEs’ adoption of IS backup systems were summarized via a literature review. The constructs included financial consideration (FC), the IS incident, and business IS decision making (BISD-M). In the second stage, an expert questionnaire was applied; an advanced hybrid modified Delphi method (MDM) and analytic hierarchy process (AHP) with expert input were constructed to identify the sorting of overall weights based on the 11 factors included in the first stage. Following the empirical conclusions, the top three critical factors were “disaster loss amount”, “enterprise’s downtime”, and “supplier’s contractual requirements”. The conclusions of this study indicated that two factors were included in the FC construct; thus, the FC construct influenced IS the most, and the BISD-M construct took second place. In the final stage, through re-checking three actual cases, the results of this study were verified with specific respect to the FC. In conclusion, to popularize IS backup systems among SMEs and fully implement IS, manufacturers may start from the FC in the hope that the severe impact caused by IS incidents featuring malicious attacks can be slowed down and the losses encountered can be lowered. The empirical results and conclusions of this study can be used for reference by SMEs, and both theoretical and empirical foundations have been provided for further studies in academic circles; the results above also show a significant application contribution of this study.

show abstract

Section: Common Virus Attack Methodsmentioning

confidence: 99%

Section: Common Virus Attack Methodsmentioning

confidence: 99%

Identification of SMEs in the Critical Factors of an IS Backup System Using a Three-Stage Advanced Hybrid MDM–AHP Model

Chen

Chou

Lin³

et al. 2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…Te features are extracted from four main categories, and the dataset is evaluated by anomaly detection patterns using various supervised machine learning algorithms. Te authors fuse web mining features and train them using four machine algorithm models-logistic regression, support vector machine, decision tree, and random forest-while extracting Websocket, Web Worker, Postmessage and message events, Messageloop events, CPU usage, keyword matching, computational functions, and dynamic functions to train the model, and the frst three features were found to be the highest identifers for identifying malicious mining sites [10]. In general, there are two forms of illegal mining, browser-based mining and binary-based malicious mining code, with the latter being stealthier and harder to detect.…”

Section: Related Workmentioning

confidence: 99%

IoT Security Detection Method Based on Multifeature and Multineural Network Fusion

Zhu

Zhang

Liu

et al. 2023

Security and Communication Networks

View full text Add to dashboard Cite

IoT security detection plays an important role in securing the IoT ecosystem. The current detection systems suffer from poor fault tolerance and inefficient detection results. To address the IoT security vulnerability, the paper designs a multifeature fusion-based IoT security detection model to simulate an attacker sending test commands to IoT nodes. Firstly, the data collection algorithm is introduced, and the collected dataset is analyzed by three neural network models, namely, RNN, LSTM, and GRU, respectively. The best scoring model is selected as the classifier to identify vulnerabilities and achieve IoT security detection.

show abstract

“…Cryptojacking is an illegal and unauthorized mining activity on the victim’s computer, using the computational power of the victim’s computer to extract cryptocurrencies, which generates large computational consumption, reducing the computational efficiency of the victim’s computer. Moreover, this attack may be used by a powerful attacker to increment their computationally power, posing a risk to any blockchain based on mining [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ].…”

Section: Introductionmentioning

confidence: 99%

On Detecting Cryptojacking on Websites: Revisiting the Use of Classifiers

Aponte-Novoa

Álvarez

Villanueva-Polanco

et al. 2022

Sensors

View full text Add to dashboard Cite

Cryptojacking or illegal mining is a form of malware that hides in the victim’s computer and takes the computational resources to extract cryptocurrencies in favor of the attacker. It generates significant computational consumption, reducing the computational efficiency of the victim’s computer. This attack has increased due to the rise of cryptocurrencies and their profitability and its difficult detection by the user. The identification and blocking of this type of malware have become an aspect of research related to cryptocurrencies and blockchain technology; in the literature, some machine learning and deep learning techniques are presented, but they are still susceptible to improvement. In this work, we explore multiple Machine Learning classification models for detecting cryptojacking on websites, such as Logistic Regression, Decision Tree, Random Forest, Gradient Boosting Classifier, k-Nearest Neighbor, and XGBoost. To this end, we make use of a dataset, composed of network and host features’ samples, to which we apply various feature selection methods such as those based on statistical methods, e.g., Test Anova, and other methods as Wrappers, not only to reduce the complexity of the built models but also to discover the features with the greatest predictive power. Our results suggest that simple models such as Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, and k-Nearest Neighbor models, can achieve success rate similar to or greater than that of advanced algorithms such as XGBoost and even those of other works based on Deep Learning.

show abstract

MinerGuard: A Solution to Detect Browser-Based Cryptocurrency Mining through Machine Learning

Cited by 6 publications

References 35 publications

Identification of SMEs in the Critical Factors of an IS Backup System Using a Three-Stage Advanced Hybrid MDM–AHP Model

Identification of SMEs in the Critical Factors of an IS Backup System Using a Three-Stage Advanced Hybrid MDM–AHP Model

IoT Security Detection Method Based on Multifeature and Multineural Network Fusion

On Detecting Cryptojacking on Websites: Revisiting the Use of Classifiers

Contact Info

Product

Resources

About