Conferences > 2024 International Conference...

Machine Learning for Fraud Detection in Blockchain Transaction

Publisher: IEEE Cite This  PDF

Kirandeep Kaur ; R. Venkatesh ; Sumit Pundir ; S Brindha ; V Arul Mary Rexy ; Ramanjeet Singh All Authors

85     
Full
Text Views

Abstract Abstract:
Fraud detection in blockchain transactions is critical as the technology becomes more integrated into financial systems.
Document Sections Traditional rule-based systems have become ineffective against new fraudulent tactics. To solve these problems, a
unique machine learning-based approach for detecting fraud in blockchain transactions is proposed in the study.
I. Introduction
Advanced machine learning techniques such as Random Forests, Support Vector Machines, and Isolation Forest are
II. Related Works used to overcome existing systems’ limitations. The data are collected meticulously and preprocessed, feature
engineering is employed, and the models are trained to improve accuracy and adaptability. Compared to existing
III. Proposed Method systems, the proposed technique outperforms with an accuracy of 0.98, precision of 0.94, recall of 0.93, and a false

IV. Results and Discussion

positive rate of 0.06. It also has an average processing time of 70 milliseconds, improving computational efficiency. The
study represents a significant step forward in blockchain transaction security, providing dynamic protection against
VI. Conclusion fraudulent activity and bolstering the integrity of decentralized financial ecosystems.

Authors Published in: 2024 International Conference on Knowledge Engineering and Communication Systems (ICKECS)

Figures
Date of Conference: 18-19 April 2024 DOI: 10.1109/ICKECS61492.2024.10616821

References Date Added to IEEE Xplore: 07 August 2024 Publisher: IEEE

Keywords  ISBN Information: Conference Location: Chikkaballapur, India

Metrics
I. Introduction
Blockchain technology has become a game changer in financial transactions, providing outstanding security and
More Like This
decentralization. Despite their inherent robustness, blockchain networks face persistent fraud threats [1].
Traditional rule-based systems, considered excellent at securing transactions, are now unable to tackle dynamic
and sophisticated fraudulent activities [2]. The lack of sufficient security measures for blockchain transactions has
generated an urgent requirement for novel strategies to improve the infrastructure. The challenges created by
fraudulent actions involving blockchain transactions are numerous. The existing work highlights the limits of
traditional rule-based systems, which need help to adapt to changing threat techniques [3]. These systems are
frequently static, requiring more flexibility and real-time responsiveness to tackle evolving risks [4] successfully.
Furthermore, itoftenproduces false positives and negatives, compromising the accuracy of transaction monitoring
Sign in to
[5]. Motivated by these problems, the study provides Continue
a unique Reading
approach for detecting fraud in blockchain
transactions using machine learning (ML) techniques. The study’s primary objective is to address existing systems’
inadequacies by leveraging ML to improve accuracy, adaptability, and real-time responsiveness. The proposed
approach attempts to revolutionize fraud detection in blockchain transactions by incorporating machine learning
algorithms such as Random Forests and Support Vector Machines (SVM). The objectives of the paper are
threefold. Firstly, it aims to introduce a paradigm change from rule-based methods to ML-driven fraud detection
tactics, addressing existing systems’ constraints. Second, it tries to outline a comprehensive methodology that
includes data collection, preprocessing, model training, real-time monitoring, and efficiency optimization. Third, it
attempts to compare the performance of the proposed System to existing approaches, demonstrating greater
accuracy, precision, recall, and lower false favorable rates.
×

Access to this document requires a subscription.

Authors 
IEEE offers both personal and institutional subscriptions. Whether
Figures 
you are an academic, a practitioner, or a student, IEEE offers a
References range of individual and institutional subscription options that can 
meet your needs.
Keywords 
Metrics LEARN MORE Close 
IEEE Personal Account Purchase Details Profile Information Need Help? Follow

CHANGE PAYMENT OPTIONS COMMUNICATIONS US & CANADA: +1 800    

USERNAME/PASSWORD PREFERENCES 678 4333
VIEW PURCHASED
DOCUMENTS PROFESSION AND WORLDWIDE: +1 732
EDUCATION 981 0060

TECHNICAL INTERESTS CONTACT & SUPPORT

About IEEE Xplore | Contact Us | Help | Accessibility | Terms of Use | Nondiscrimination Policy | IEEE Ethics Reporting  | Sitemap |
IEEE Privacy Policy
A public charity, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity.

© Copyright 2025 IEEE - All rights reserved, including rights for text and data mining and training of artificial intelligence and similar technologies.