Hybrid Machine Learning Model for Detecting Bangla Smishing Text Using BERT and Character-Level CNN

2025-02-03

AI Summary

To address the surging security and financial threats posed by Bangla-language smishing (SMS phishing), this paper proposes a fine-grained three-class detection model that accurately distinguishes normal, promotional, and fraudulent messages. Methodologically, we introduce a novel hybrid architecture integrating BERT-based contextual embeddings with character-level CNNs, augmented by an attention mechanism to emphasize discriminative textual segments—thereby overcoming the limitations of conventional binary classification. The model is trained end-to-end within a multiclass classification framework. Evaluated on a dedicated Bangla smishing dataset, it achieves an overall accuracy of 98.47%, with high precision (98.21%) and recall (97.89%) specifically for fraudulent instances—substantially outperforming traditional classifiers such as SVM and XGBoost. This work establishes a reproducible, high-performance paradigm for fine-grained phishing detection in low-resource language settings.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Language and Vision

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Smishing is a social engineering attack using SMS containing malicious content to deceive individuals into disclosing sensitive information or transferring money to cybercriminals. Smishing attacks have surged by 328%, posing a major threat to mobile users, with losses exceeding $54.2 million in 2019. Despite its growing prevalence, the issue remains significantly under-addressed. This paper presents a novel hybrid machine learning model for detecting Bangla smishing texts, combining Bidirectional Encoder Representations from Transformers (BERT) with Convolutional Neural Networks (CNNs) for enhanced character-level analysis. Our model addresses multi-class classification by distinguishing between Normal, Promotional, and Smishing SMS. Unlike traditional binary classification methods, our approach integrates BERT's contextual embeddings with CNN's character-level features, improving detection accuracy. Enhanced by an attention mechanism, the model effectively prioritizes crucial text segments. Our model achieves 98.47% accuracy, outperforming traditional classifiers, with high precision and recall in Smishing detection, and strong performance across all categories.

Problem

• Machine Learning
• Bangla SMS Fraud
• Classification

Innovation

• Hybrid Machine Learning Model
• BERT and Character-level CNN
• Bangla SMS Fraud Detection