Abstract :

The Proliferation Of Mobile Communication Technologies Has Led To An Exponential Increase In Short Message Service (SMS) Spam, Posing Significant Security Threats And User Inconvenience. Traditional Spam Filtering Techniques Have Become Increasingly Ineffective Against Sophisticated Evasive Methods Employed By Modern Spammers. This Research Presents A Comprehensive Investigation Into Advanced Machine Learning Approaches For Detecting And Analyzing Evasive SMS Spam Techniques. The Proposed System Implements An Ensemble-based Machine Learning Framework That Combines Multiple Classification Algorithms Including Random Forest, Support Vector Machines, Logistic Regression, Gradient Boosting, And Naïve Bayes. The System Incorporates Advanced Feature Engineering Techniques That Go Beyond Traditional Text Analysis, Including URL Obfuscation Detection, Phone Number Pattern Recognition, Urgency Indicator Analysis, And Linguistic Feature Extraction. A Comprehensive Dataset Of 15,000 SMS Messages Was Created, Containing Both Legitimate (ham) And Spam Messages With Sophisticated Evasion Techniques. The System Achieved A Remarkable 96.8% Accuracy In Spam Detection With A False Positive Rate Of Only 1.2%. The Ensemble Model Demonstrated Superior Performance Compared To Individual Classifiers, Particularly In Identifying Sophisticated Evasion Strategies.The Implemented Web Application Provides Real-time Spam Analysis With Interactive Visualizations, Risk Assessment Metrics, And Comprehensive Reporting Features. The System Architecture Supports Multi-user Access With Separate Administrative And User Interfaces, Enabling Efficient Management Of Datasets, Model Training, And Prediction Monitoring. This Research Contributes To The Field Of SMS Security By Providing A Robust, Scalable Solution That Effectively Counters Modern Spam Evasion Techniques While Maintaining High Usability And Real-time Performance. The Findings Demonstrate That Ensemble Machine Learning Approaches, When Combined With Comprehensive Feature Engineering, Can Significantly Enhance Spam Detection Capabilities In The Evolving Landscape Of Mobile Communication Threats. Keywords: SMS Spam Detection, Machine Learning, Ensemble Methods, Evasive Techniques, Natural Language Processing, Cybersecurity, Realtime Analysis, Feature Engineering

Published:

06-3-2026

Issue:

Vol. 26 No. 3 (2026)

Page Nos:

26-31

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite