Classification of credit card holders based on random forest algorithm

Lei, Guowei; Su, Sunqing; Liao, Wenliang

doi:10.1145/3453800.3453806

Cited by 2 publications

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study also examined how data provenance and quality affect how well data mining-based IDSs work. In particular, we focus on supervised data mining models deployed on historical labelled data where the analysis was on Recurrent Neural Network (Tsantekidis et al, 2022;Yin et al, 2017), Decision Tree (Fan, 2021), (Lei et al, 2021) Random Forest, Support Vector Machines (SVM) (Schlag et al, 2021), Logistic Regression (Zhang & Qin, 2022), and Naïve Bayes (Guo, 2022). This article is structured into four sections, beginning with literature studies in the related area, followed by the study's methodology.…”

Section: Introductionmentioning

confidence: 99%

Enhanced Network Security: A Data Mining Approach to Intrusion Detection

Teoh Chun Hwung,

Yusof

2024

jdsd

View full text Add to dashboard Cite

In the ever-evolving landscape of network security, the role of Intrusion Detection Systems (IDSs) is critical. These systems serve as the guardians of digital networks, defending against a relentless tide of cyber threats. However, the efficacy of IDSs in accurately detecting and preventing these threats remains a critical concern. The emergence of new attack vectors and the growing sophistication of cyberattacks underscore the need for innovative approaches to intrusion detection. Nonetheless, there is an information vacuum about the efficacy of different data mining approaches in intrusion detection. This study investigates the deployment of seven data mining models, which include Decision Tree, Random Forest, Support Vector Machine (SVM), Logistic Regression, Naïve Bayes, and two types of Recurrent Neural Networks (RNN) on five (5) IDS benchmark datasets. Based on the undertaken experiments, it is evident that the RNN is the best classifier as it obtained 100% accuracy for all datasets. Due to its strength in modelling time-dependent and sequential data problems, RNN has become powerful in predicting future attacks. As our digital world evolves, so must our cyber defences; hence, this study strives to equip network security professionals with the knowledge and tools needed to fortify their networks, ensuring resilience against the ever-adaptive landscape of cyber threats.

show abstract