Comparative Detection of SQL Injection Attacks Using Machine Learning and a Convolutional Neural Network

This preprint presents a comparative study of machine learning and deep learning methods for detecting SQL injection attacks through binary classification of SQL-query text. It compares logistic regression, random forest, support vector machine, Naive Bayes, decision tree, and a one-dimensional convolutional neural network using accuracy, precision, recall, and F1 score. The manuscript is derived from the author’s undergraduate thesis completed at Chittagong University of Engineering and Technology (CUET) in 2023, under the supervision of Professor Dr. Jamal Uddin Ahamed. This version summarizes the results reported in the thesis and explicitly documents limitations, including missing dataset provenance and experimental settings, inconsistent CNN descriptions, and discrepancies between result tables. These details require verification against the original implementation. This manuscript has not undergone peer review.