Machine-Learning Techniques for Predicting Phishing Attacks in Blockchain Networks: A Comparative Study

Joshi, Kandarp; Bhatt, Chintan; Shah, Keval; Parmar, D. J.; Corchado, Juan M.; Bruno, Alessandro; Mazzeo, Pier Luigi

doi:10.3390/a16080366

Cited by 14 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zero-trust policies are gradually being introduced as a method, but their deployment is still ongoing and requires a considerable amount of time. A more accurate approach to phishing attack detection involves the use of machine learning models with specific features to automatically classify attempts as phishing attacks or legitimate ones [135]. In the context of preventing phishing attacks in blockchain, federated decision trees may be employed to consolidate information from various miners or nodes, collaboratively constructing a model capable of identifying phishing attacks.…”

Section: Discussionmentioning

confidence: 99%

Decision Tree-Based Federated Learning: A Survey

Wang,

Gai

2024

Blockchains

View full text Add to dashboard Cite

Federated learning (FL) has garnered significant attention as a novel machine learning technique that enables collaborative training among multiple parties without exposing raw local data. In comparison to traditional neural networks or linear models, decision tree models offer higher simplicity and interpretability. The integration of FL technology with decision tree models holds immense potential for performance enhancement and privacy improvement. One current challenge is to identify methods for training and prediction of decision tree models in the FL environment. This survey addresses this issue and examines recent efforts to integrate federated learning and decision tree technologies. We review research outcomes achieved in federated decision trees and emphasize that data security and communication efficiency are crucial focal points for FL. The survey discusses key findings related to data privacy and security issues, as well as communication efficiency problems in federated decision tree models. The primary research outcomes of this paper aim to provide theoretical support for the engineering of federated learning with decision trees as the underlying training model.

show abstract

Section: Discussionmentioning

confidence: 99%

Decision Tree-Based Federated Learning: A Survey

Wang,

Gai

2024

Blockchains

View full text Add to dashboard Cite

show abstract

“…This research employed well-established evaluation metrics, including accuracy, fmeasure, the area under the curve (AUC), and Matthew's correlation coefficient (MCC), to assess and contrast the predictive capabilities of the various investigated models. The selection of these performance indicators was based on their frequent utilization in prior studies in the assessment of rule-based and ML-based software risk prediction models [43][44][45]. Furthermore, these metrics are reported to be dependable collectively, as they consider all areas of the confusion matrix produced for each developed model [46,47].…”

Section: Methodsmentioning

confidence: 99%

Software Requirement Risk Prediction Using Enhanced Fuzzy Induction Models

Mamman,

Balogun,

Basri

et al. 2023

Electronics

View full text Add to dashboard Cite

The development of most modern software systems is accompanied by a significant level of uncertainty, which can be attributed to the unanticipated activities that may occur throughout the software development process. As these modern software systems become more complex and drawn out, escalating software project failure rates have become a critical concern. These unforeseeable uncertainties are known as software risks, and they emerge from many risk factors inherent to the numerous activities comprising the software development lifecycle (SDLC). Consequently, these software risks have resulted in massive revenue losses for software organizations. Hence, it is imperative to address these software risks, to curb future software system failures. The subjective risk assessment (SRM) method is regarded as a viable solution to software risk problems. However, it is inherently reliant on humans and, therefore, in certain situations, imprecise, due to its dependence on an expert’s knowledge and experience. In addition, the SRM does not allow repeatability, as expertise is not easily exchanged across the different units working on a software project. Developing intelligent modelling methods that may offer more unbiased, reproducible, and explainable decision-making assistance in risk management is crucial. Hence, this research proposes enhanced fuzzy induction models for software requirement risk prediction. Specifically, the fuzzy unordered rule induction algorithm (FURIA), and its enhanced variants based on nested subset selection dichotomies, are developed for software requirement risk prediction. The suggested fuzzy induction models are based on the use of effective rule-stretching methods for the prediction process. Additionally, the proposed FURIA method is enhanced through the introduction of nested subset selection dichotomy concepts into its prediction process. The prediction performances of the proposed models are evaluated using a benchmark dataset, and are then compared with existing machine learning (ML)-based and rule-based software risk prediction models. From the experimental results, it was observed that the FURIA performed comparably, in most cases, to the rule-based and ML-based models. However, the FURIA nested dichotomy variants were superior in performance to the conventional FURIA method, and rule-based and ML-based methods, with the least accuracy, area under the curve (AUC), and Mathew’s correlation coefficient (MCC), with values of approximately 98%.

show abstract

“…Deep learning (DL) algorithms exhibit potential superiority over ML counterparts, as evidenced by the exceptional performance of models like bidirectional encoder representations from transformers (BERT) and long short-term memory (LSTM) in phishing detection [25]. However, not all DL algorithms fare better, as demonstrated by Joshi et al [26], highlighting the need for a nuanced comparison against other methodologies like Random Forest (RF) and Extreme Gradient Boost (XG Boost).…”

Section: Related Workmentioning

confidence: 99%

Phishing Attacks among Internet Banking Users in Nigeria: An Exploration of Remedial Strategies

Tepede Dipo

2024

IJLTEMAS

View full text Add to dashboard Cite

This study meticulously examines phishing attacks targeting Nigerian internet banking users. It delves into prevailing trends, methodologies, and the effectiveness of existing countermeasures. It also proposes robust, tailored security measures in the form of a conceptual model. Similarly, by analyzing the latest tools and techniques, the study emphasizes the need for adaptive and continuously learning anti-phishing systems. The outcomes aim to empower stakeholders in the Nigerian banking sector, including policymakers, financial institutions, and users, to combat the complex and evolving threats posed by phishing attacks.

show abstract

Machine-Learning Techniques for Predicting Phishing Attacks in Blockchain Networks: A Comparative Study

Cited by 14 publications

References 24 publications

Decision Tree-Based Federated Learning: A Survey

Decision Tree-Based Federated Learning: A Survey

Software Requirement Risk Prediction Using Enhanced Fuzzy Induction Models

Phishing Attacks among Internet Banking Users in Nigeria: An Exploration of Remedial Strategies

Contact Info

Product

Resources

About