EAI Endorsed Transactions on Scalable Information Systems

EcomFraudEX: An Explainable Machine Learning Framework for Victim-Centric and Dual-Sided Fraud Incident Classification in E-Commerce

Salman Farsi — 2025-01-20

The popularity of e-commerce businesses and online shopping is experiencing rapid growth all around the world. Nowadays, people are more inclined to shop online than in the actual shops. Due to this advancement, fraudsters have set new traps to deceive consumers. Whether it is true that customers often become victims of fraud, it also happens that a fraud customer tries to deceive the seller and hassle the seller intentionally in several ways. To address these issues, an automated system is required so that fraud incidents can be classified. This will facilitate taking legal action and reporting to consumer rights authorities. Existing research on fraud detection and prevention didn't cover customer and seller-side fraud simultaneously. Besides, most of the work focused on fraud detection rather than post-fraud incident classification. To overcome these gaps, this research endeavor conducts a thorough online survey of customers and sellers to gather incident-specific victim data on fraud cases and it addresses the issue for both customer and seller. This paper proposes a machine learning (ML) based explainable fraud incident classification framework EcomFraudEX, that can efficiently classify these fraud incidents and analyze the reason behind each incident. This framework particularly focuses on proper feature selection techniques, hyper-parameter tuning of models, and exploring different ML and ensemble models. Ensemble majority voting schemes consisting of Random Forest (RF), XGBoost, and CatBoost achieved the highest F1-score of 96% with the Chi-Square feature selection technique in the customer complaint dataset and 98% with the RF feature selection technique in the seller complaint dataset. To explain the incident reasoning, Local Interpretable Model Agnostic Explanation (LIME) and Shapely Additive Explanation (SHAP) were further utilized. The proposed scheme achieved a 1.57% higher F1-score and 2.13% higher accuracy than previous works.

Design and Simulation of Quasi-microstrip Yagi Antenna in Railway Mobile Communication

Junhua Shao — 2025-01-16

PROBLEM STATEMENT: The performance and size constraints of the current railway mobile communication systems force the development of more effective communication methods.

OBJECTIVES: This study aims to build a quasi-microstrip Yagi antenna with a 900MHz centre frequency that minimizes the antenna's physical dimensions and improves communication capabilities between trains.

METHODS: HFSS simulation was used during the design phase to optimize the antenna's characteristics, which included bending the active oscillator to increase gain. Important performance indicators were assessed, including voltage standing wave ratio (VSWR), bandwidth, and return loss.

RESULTS: The optimized antenna produced a VSWR of less than 2, a maximum gain of 8.35dB, a bandwidth of 105MHz (spanning from 845MHz to 950MHz), and a return loss of -19.49dB at the centre frequency. According to these results, the antenna satisfies the operating criteria for railway mobile communication.

CONCLUSION: The quasi-microstrip Yagi antenna proves useful in engineering applications since it not only meets the communication requirements of railway systems but also shrinks considerably in size compared to conventional Yagi antennas.

Exploring Social Media Research Trends in Malaysia using Bibliometric Analysis and Topic Modelling

Rehan Tariq — 2025-01-02

This study explores the evolving dynamics of social media research in Malaysia. The main objective is to identify trends and patterns in research, specifically examining the volume and focus of scholarly articles over the last decade. Using bibliometric analysis and topic modelling, the study identifies major research clusters and key themes like digital marketing, political communication, and public health, and to map out collaborations among researchers. The findings show a significant increase in social media-related studies, highlighting a trend towards more varied and complex topics. This includes a greater emphasis on social media's role in political communication, consumer behaviour, and crisis management. Looking forward, this study suggests that future studies should explore the applications of emerging technologies such as artificial intelligence (AI) on social media practices, assess the spread and impact of negative information such as fake news and hate speech, and extend cross-disciplinary methodologies to fully understand the extensive effects of social media.