This dataset is taken from Kaggle: Cybersecurity Intrusion Detection Dataset by dnkumars.
The project aims to detect cyberattacks in network traffic using both technical features and user behavioral data.
The dataset includes information about network packets and user activity. Each record represents a single network connection or user session, described by several numerical and categorical features.
The classification model predicts the variable attack_detected, with values:
- 1 - attack / suspicious activity detected
- 0 - normal activity
The analysis includes data preprocessing, dimensionality reduction using PCA, UMAP, and t-SNE, and classification using Logistic Regression, SVM, k-NN, Random Forest and a Neural Network (Keras Sequential), with evaluation based on accuracy, precision, recall, and F1-score.