The UNSW-NB15 dataset plays a crucial role in advancing research on network intrusion detection. It is primarily used to evaluate the effectiveness of machine learning models, such as Intrusion Detection Systems (IDS), in identifying and classifying different types of network attacks.
These attack categories include:
- Probe
- R2L (Remote to Local)
- DoS (Denial of Service)
- U2R (User to Root)
The dataset consists of 49 features capturing various aspects of network traffic, enabling model development and thorough experimentation. This project focuses on leveraging the UNSW-NB15 dataset to enhance intrusion detection and explore anomaly detection strategies for bolstering network security.
- Proficiency in analyzing and interpreting network traffic data.
- Ability to generate and recognize attack signatures and patterns.
- Enhanced knowledge of network protocols and security vulnerabilities.
- Experience with machine learning model building, evaluation, and tuning.
- pandas – data manipulation and analysis
- numpy – numerical computations
- seaborn – statistical data visualization
- matplotlib – plotting static and interactive visualizations
- scikit-learn (sklearn) – machine learning algorithms, preprocessing, and evaluation
- XGBoost – efficient and scalable gradient boosting library
- Heatmaps – for visualizing relationships and correlations between variables
- Download the UNSW-NB15 dataset from Kaggle.
- Extract the dataset and explore the 8 classified
.csvfiles. - Download or clone the Python code repository for this project.
- Run the code and analyze the results and visualizations produced.
