---
tags: netsec
title: Netsec report final
---

# Resources

Report: [link](https://www.overleaf.com/1328617962prtvcqpnjmvc)

Slides: TBA

## Dataset:

We use the following datasets:
- Bot-Iot
- [Source](https://research.unsw.edu.au/projects/bot-iot-dataset)
- [Paper](https://arxiv.org/abs/1811.00701)
- Collab
- UNSW-NB15
- [Source](https://research.unsw.edu.au/projects/unsw-nb15-dataset)
- [Paper](https://www.researchgate.net/publication/287330529_UNSW-NB15_a_comprehensive_data_set_for_network_intrusion_detection_systems_UNSW-NB15_network_data_set)
- [Collab](https://colab.research.google.com/drive/11nT0q0Yp9JyVZrea8XyrOp8fxREUSlHK?authuser=1#scrollTo=roenEnxb7nn9)

## Feature selection:
- Filter-based / statistical based on chi2 - [link](https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectKBest.html)
- PCA for dimensionality reduction, to select top components - [link](https://scikit-learn.org/stable/modules/generated/sklearn.decomposition.PCA.html)

## Models:
- Logistic Regression
- Support Vector Machine (SVM)
- Multi-layer Perceptron / DNN
- XGBoost
- Tentative, +- more based on initial tests - other candidates could include - Naive Bayes Classifier, GA-ANFIS (Soft computing ML)