Ml Anomaly Detection

Detection of network traffic anomalies using unsupervised machine learning
Alternatives To Ml Anomaly Detection
Project NameStarsDownloadsRepos Using ThisPackages Using ThisMost Recent CommitTotal ReleasesLatest ReleaseOpen IssuesLicenseLanguage
100 Days Of Ml Code17,892
a year ago9mitJupyter Notebook
6 days ago6mitPython
Transfer learning / domain adaptation / domain generalization / multi-task learning etc. Papers, codes, datasets, applications, tutorials.-迁移学习
Stanford Cs 229 Machine Learning10,399
3 years ago11mit
VIP cheatsheets for Stanford's CS 229 Machine Learning
Awesome Artificial Intelligence7,363
a month ago39
A curated list of Artificial Intelligence (AI) courses, books, video lectures and papers.
Anomaly Detection Resources6,883
a month ago10agpl-3.0Python
Anomaly detection related books, papers, videos, and toolboxes
Pyod6,845331a day ago83July 05, 2022160bsd-2-clausePython
A Comprehensive and Scalable Python Library for Outlier Detection (Anomaly Detection)
Mlxtend4,283956023 days ago49May 27, 2022128otherPython
A library of extension and helper modules for Python's data analysis and machine learning libraries.
Awesome Community Detection2,111
3 days ago1cc0-1.0Python
A curated list of community detection research papers with implementations.
2 years ago18agpl-3.0Python
The standard package for machine learning with noisy labels and finding mislabeled data. Works with most datasets and models.
Karateclub1,84337 days ago106June 04, 20222gpl-3.0Python
Karate Club: An API Oriented Open-source Python Framework for Unsupervised Learning on Graphs (CIKM 2020)
Alternatives To Ml Anomaly Detection
Select To Compare

Alternative Project Comparisons

[1] Overview

This project was done in the subject, COMP90073 (Security Analytics) taken in Semester2, 2020 in the University of Melbourne.


[2] Features

More details in the anomaly_detection_reports.pdf

  • Feature1: Numeric value (existing + newly generated) + Standardscaler + PCA

  • Feature2: Feature1 + One-hot encoded categorical feature

  • Feature3: Scale (Cumulative features grouped by stream_id + time-based feature) + PCA

[3] Model

  1. Iforest
  2. OneclassSVM

[4] Hyperparameter tuning (2 examples among 6)

  • Criteria of setting a threshold: Accuracy > 0.88 and Max(TPR-FPR)
  1. OCSVM + feature3

alt text

  1. Iforest + feature 3

alt text

[5] Clustering visualisation and Evaluation (2 examples among 6)

  1. OCSVM + feature3


alt text


alt text

  1. Iforest + feature3


alt text


alt text

[6] Interpretation of the result

alt text alt text

  • Attack Timeline alt text

[7] Generating Adversarial samples (FGSM)

  • FGSM generates adversarial samples with the error rate of almost 100%.

alt text

Popular Machine Learning Projects
Popular Unsupervised Learning Projects
Popular Machine Learning Categories
Related Searches

Get A Weekly Email With Trending Projects For These Categories
No Spam. Unsubscribe easily at any time.
Machine Learning
Deep Learning
Unsupervised Learning
Anomaly Detection
Network Analysis
Adversarial Learning