Enhancing Security of Mobile Payment Apps using Context-Aware Anomaly Detection
Scope
Context-aware anomaly detection evaluates user behavior and transaction context in real-time to improve the security of mobile payment apps. The demand for mobile payment apps to have better security is highlighted by the recent increase in fraud and cyberattack incidents. By the consideration of many contextual factors such as user location, transaction history, and device information, context-aware anomaly detection provides a proactive method to detect the suspicious activity. Context-aware anomaly detection techniques employ machine learning algorithms to assess huge amounts of data in real-time and identify anomalies from typical user behavior or transaction patterns, potentially discovering security risks. One of these methods is risk scoring, which provides an extensive evaluation of high-risk transactions by assigning transaction risk levels based on contextual elements. Machine learning models use both supervised and unsupervised learning to identify abnormal behavior. Risk scoring employs contextual aspects to assign transaction risk levels, allowing for a deeper analysis of high-risk transactions.
In order to improve security, mobile payment apps are rapidly using AI and machine learning for real-time anomaly detection, including biometric authentication techniques like fingerprint and facial recognition. In an effort to increase transaction security and transparency, certain apps are looking into blockchain technology. Further developments in AI will lead to more advanced anomaly detection systems that can recognize complex fraud patterns. More contextual data will be provided by integrating with Internet of Things (IoT) devices, improving security even more. Additionally, by utilizing methods like federated learning to train models without exposing sensitive data, future mobile payment apps might prioritize customer privacy first. Despite its potential it has several obstacles such as data privacy concerns, false positives and adversarial attacks.
It might be difficult to protect user privacy when gathering and evaluating sensitive data, but strong encryption and anonymization strategies can aid in this attempt. Algorithms can be continuously improved and user feedback can be integrated to reduce high false positive rates, which can damage user trust. Resilient anomaly detection techniques must be used to prevent adversarial attacks, when hackers change user behavior to avoid detection. Real-time analysis of user behavior and transaction context by context-aware anomaly detection holds great promise to improve the security of mobile payment apps. These apps can identify and stop fraudulent activity with more accuracy, offering users a safer and secure transaction experience, using machine learning algorithms and sophisticated contextual data. To effectively use this technology's potential, however, issues like false positives and data privacy concerns must be resolved.
The topics relevant to this special issue include but are not limited to:
- Role of context-aware anomaly detection to enhance mobile payment security.
- Enhancing mobile payment security with anomaly detection.
- Advancements in machine learning algorithms to safeguard the mobile payment transaction.
- Analysis of contextual insights in mobile payment anomaly detection.
- Strategies to overcome the threat landscape using context-aware anomaly detection.
- Implementing biometric authentication to improve mobile payment app security.
- Applications of federated learning in mobile payment apps to prioritize user privacy.
- Exploring blockchain technology for enhanced transaction security in mobile payments.
- Challenges and opportunities in mobile payment innovation for balancing security and transparency.
- Employing IoT devices for contextual insights in mobile payment security.
Special Issue Timeline:
Submission Deadline: 11.11.2024
Author Notification: 21.01.2025
Revised Submission: 31.03.2025
Final Acceptance: 01.07.2025
Main Guest Editor:
Dr. Mehdi Gheisari, Department of Computer Science, Islamic Azad University, Tehran, Iran. Email: mehdi.gheisari61@gmail.com, mehdigheisarics@outlook.com
Guest Editors:
Dr. Subhendu Kumar Pani, Krupajal Engineering College, Biju Patnaik University of Technology, India. Email: pani.cse@oec.ac.in
Dr. Kamalakanta Muduli, Department of Mechanical Engineering, The Papua New Guinea University of Technology, Papua New Guinea. Email: kamalakanta.muduli@pnguot.ac.pg
