cyber-security - Machine Learning for Network Intrusion Detection &

项目说明

Cyber Security

> Cyber Security related self-learnings. _CMPT 318: Cyber Security_ can be found Here

Projects Here

Network Intrusion Detection

* KDD cup ‘99
* 100% Accuracy
* Blog post
* NSL-KDD (corrected dataset based off KDD cup ‘99’)
* 98% Accuracy
* Effects of Dimensionality Reduction
* Blog post
* UNSW_NB15
* 94.8% Accuracy
* 91% Accuracy with a feedforward neural network, 94.8% with Ensemble (of non neural network techniques, '')
* Autoencoder of Anomaly Detection
* Upsupervised with PYOD, which is a "A Python Toolbox for Scalable Outlier Detection (Anomaly Detection)"

Financial Anomaly Detection

* SPX 500 pointing out _oddities_ from the past in an unsupervised manner
* Some argue the stock market is a random signal, which tends to have a positi in the *long run*. So, I tried Anomaly Detection on it. Perhaps it's an *extream* bull run and bear run detector

Utilities

* Rainbow Table of several Hashing Algorithms* A simple hash lookup table
* Bandwidth Hog
* Download a file many times, but do not save the data. Useful for stressing out a network... possibly your own
* File (de)Encryption
* boilerplate code for my latter use
* unExpectedProcessChecker
* outputs the mutually exclusive set of processes, given set A, which is hardcoded, and set B, which is from the windows en being run on
* Simple keylogger
* Dos script

Related projects

1. CMPT 318: Cyber Security
* Presentation on Anomaly Detection
2. myPybackdoor
* Na Directories
* Download, Create, and Delete Files
* Execute custom CMD commands
* Get wifi Password list
3. chromePasswordThief
* Grab passwords saved in chrome and email them out
* Blog post
4. bandwidth-hog
* Download a file many times, but do not save the data. Useful for stressing out a network... possibly your own
5. Rainbow table of ser algorithm

* A simple hash lookup table, easily expandable to fit your needs

项目信息

开发语言：Jupyter Notebook

Star 数：225

技术要点

该项目属于病毒分析方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕Machine Learning for Network Intrusion Detection & Misc Cyber Security Utilities 展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。