A Dual-Pipeline Imbalance-Robust Framework for SMS Spam Detection: Achieving Flawless Precision via SMOTE-Augmented Ensembles with Rigorous Statistical Validation
Jul 2026· Journal Computer and Technology· Vol 4, pp. 107-122· 0 citations· 22 references
TL;DR
This study presents a rigorous dual-pipeline machine learning framework that systematically addresses the challenges of class imbalance in statistical text mining and establishes a highly scalable, mathematically verified, and low-latency solution suitable for integration into real-time telecom filtering gateways.
Abstract
The rapid proliferation of digital communication has exponentially increased the volume of Short Message Service (SMS) spam, exposing mobile users to systemic convenience disruptions, productivity drops, and severe financial losses through sophisticated fraudulent schemes. To construct a highly dependable filtering mechanism, this study presents a rigorous dual-pipeline machine learning framework that systematically addresses the challenges of class imbalance in statistical text mining. Utilizing a verified dataset of 5,572 Indonesian-context short messages, the raw textual corpus is subjected to uniform case normalization, structural URL extraction, and character filtering before feature projection via Term Frequency–Inverse Document Frequency (TF-IDF) vectorization. To overcome the inherent accuracy paradox of skewed class distributions, the experimental design evaluates a baseline pipeline (imbalanced data) against a synthetic data augmentation pipeline leveraging the Synthetic Minority Oversampling Technique (SMOTE) across four distinct classifiers: Logistic Regression, Naive Bayes, Linear Support Vector Machine (Linear SVM), and Random Forest. Empirical results demonstrate that while the baseline Linear SVM serves as the optimal standalone model for overall balance, achieving a peak accuracy of 98.11% and a dominant F1-Score of 92.83%, the SMOTE-augmented Random Forest configuration yields an exceptional high-security alternative by securing a flawless 100.00% precision envelope alongside an 83.89% recall rate. Advanced post-hoc evaluations including McNemar's statistical significance tests (, for Random Forest), qualitative error analyses of semantic edge cases, and runtime profiling confirm that the developed architecture establishes a highly scalable, mathematically verified, and low-latency solution suitable for integration into real-time telecom filtering gateways.
A high-efficiency detection framework utilizing DistilBERT, a distilled knowledge representation of the BERT transformer is proposed, substantiate the viability of Knowledge Distillation as a mechanism to deploy state-of-the-art semantic security filters on edge infrastructure.
Mrinal Mrinal, Neeraj Kumar· International Journal of Cre...· 0 citations
The rapid evolution of adversarial spam within resource-constrained enterprise networks requires an urgent transition away from static heuristic filters. While deep learning transformers provide high accuracy, their heavy computational footprints require expensive hardware acceleration (GPUs) that remains impractical for local edge server deployment. This study bridges this gap by developing a low-overhead, CPU-bound Bootstrap Aggregated (Bagging) Ensemble architecture that fuses the probabilistic throughput of Multinomial Naïve Bayes (MNB) with the high-dimensional geometric separation of Support Vector Machines (SVM). Preprocessed via a rigorous natural language processing pipeline and a sub-linear TF-IDF feature mapping, evaluated against a composite benchmark corpus (N=10,000, 52% ham / 48% spam) combining Enron, SpamAssassin, and institutional logs. The hybrid engine matches the classification precision of optimized lightweight deep transformers (e.g., DistilBERT) while cutting CPU inference latency from 412.5 ms to an ultra-low 21.1 ms, the hybrid engine was evaluated against a 2024–2026 real-world benchmark corpus. Empirical results show the proposed model achieves a verified classification accuracy of 98.42% and an $F_1$-score of 98.39%—matching deep learning precision boundaries while drastically reducing mean CPU inference latency from 412.5 ms to an ultra-low 21.1 ms. This framework establishes an efficient, resource-resilient defense baseline aligned with the NIST Cybersecurity Framework (CSF) 2.0 standards for securing constrained edge environments.
Hamza Audi Giade, A. Tukur, Y. Chindo et al.· International journal of res...· 0 citations
Experimental evaluations on the UCI SMS Spam Collection dataset demonstrate that the TF-IDF-PCA-DT pipeline achieves a detection accuracy of 99% while reducing model size by 77% and inference time by 75%, demonstrating the practicality of the proposed framework for real-world deployment in resource-constrained environments.
Baraa Tareq Hammad, Ismail Taha Ahmed, Mohamed A. Hafez et al.· De Computis· 0 citations
Experimental results on English and Korean SMS datasets indicate that LLMs are highly robust for spam detection and can provide interpretable rationales by leveraging their language understanding capabilities, suggesting that LLMs are a highly effective and practical option for modern SMS spam filtering.
Minseok Hur, Sooyon Seo, Jaeho Hwang et al.· International Journal of Inf...· 0 citations
Email communication is one of the most significant modes of digital interaction, playing a central role in business communication, personal correspondence, and information exchange across the modern world. The increasing volume of unsolicited, deceptive, and malicious emails, commonly known as spam, has made accurate spam detection a challenging task for email service providers, organizations, and individual users. Traditional filtering methods primarily depend on static rule-based approaches and manually curated blacklists, which often fail to adapt to rapidly evolving spam patterns and result in inaccurate classification. Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML) have enabled the development of intelligent systems capable of analyzing email content and generating accurate spam classifications using data-driven approaches digital communication. The proposed system utilizes textual features such as word frequency, message structure, and content patterns to classify emails as spam or legitimate (ham) through supervised learning techniques.
The email dataset undergoes preprocessing procedures, including text cleaning, tokenization, stop-word removal, stemming, feature extraction using TF-IDF vectorization, and train-test splitting before model training. The predictive model is developed using Python and Scikit-learn, while Flask is employed to create a web-based interface that allows users to enter email content and obtain real-time spam classification.
M. K, S. Nandhini· International Journal of Cre...· 0 citations
Social media platforms have become major vectors for financial and cryptocurrency fraud, resulting in substantial economic losses and eroding user trust. This research presents a dual model system integrating machine learning-based user verification with deep learning-based content analysis to detect fraudulent activity more effectively than traditional single dimensional approaches. The core innovation lies in fusing these two modalities using a logical OR strategy. This design facilitates the detection of hybrid fraud schemes such as compromised legitimate accounts posting deceptive content or fake accounts spreading benign-looking messages that typically evade isolated detection systems. For user verification, ensemble methods were evaluated on a large, balanced dataset of social media profiles. Among the individual classifiers evaluated, the random forest classifier achieved the strongest performance and was selected for the final architecture due to its optimal balance of accuracy, interpretability, and computational efficiency. For content analysis, a convolutional neural network (CNN) trained on a substantial corpus of crypto-related posts demonstrated high accuracy, outperforming traditional keyword-based and recurrent neural network baselines. Ultimately, combining these models yields a system that flags significantly more fraudulent posts than either component alone. The user verification component retains interpretability through feature importance measures, while the CNN-based content analysis component operates as a less transparent classifier; the combined system can be adapted across diverse social media platforms.
B. Bokolo, Qingzhong Liu· Electronics· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
