# Image-Based Telecom Fraud Detection Method Using an Attention Convolutional Neural Network

## TL;DR

This paper introduces a new method using a convolutional neural network to detect telecom fraud more effectively, even when fraud data is extremely rare.

## Contribution

A novel attention CNN with Focal Loss is proposed to handle imbalanced telecom fraud detection.

## Key findings

- The proposed method achieves a recall of 0.7850 in detecting rare fraud cases.
- The model outperforms existing methods with an AUC of 0.8662 on real-world data.

## Abstract

In recent years, telecom fraud remains prevalent in many regions, severely impacting people’s daily lives and causing substantial economic losses. However, previous research has mainly relied on expert knowledge for feature engineering, which lags behind and struggles to adapt to the continuously evolving patterns of fraud effectively. In addition, the extreme imbalance in fraud amounts within real communication data hinders the development of deep learning methods. In response, we propose a feature transformation method to represent users’ communication behavior as comprehensively as possible, and develop a convolutional neural network (CNN) with a Focal Loss function to identify rare fraudulent activities in highly imbalanced data. Experimental results on a real-world dataset show that, under conditions of severe class imbalance, the proposed method significantly outperforms existing approaches in two key metrics: recall (0.7850) and AUC (0.8662). Our work provides a new approach for telecommunication fraud detection, enabling the effective identification of fraudulent numbers.