https://github.com/muayyad-alsadi/flowvis
visualize netflows using t-SNE
https://github.com/muayyad-alsadi/flowvis
botnets netflow
Last synced: about 1 year ago
JSON representation
visualize netflows using t-SNE
- Host: GitHub
- URL: https://github.com/muayyad-alsadi/flowvis
- Owner: muayyad-alsadi
- License: apache-2.0
- Created: 2017-07-26T19:12:16.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2017-07-26T19:27:50.000Z (almost 9 years ago)
- Last Synced: 2025-02-16T20:14:52.009Z (over 1 year ago)
- Topics: botnets, netflow
- Language: JavaScript
- Size: 172 KB
- Stars: 3
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# FlowVis
Visualize netflows using t-SNE
## Input
Prepare you netflows in the form of CSV having the following columns
```
StartTime,Dur,Proto,SrcAddr,Sport,Dir,DstAddr,Dport,State,sTos,dTos,TotPkts,TotBytes,SrcBytes,Label
```
If you don't have labels put any thing as last column like `NA` or `OTHER` or `None`
You can use any file that have the above format like
* [CTU-Malware-Capture-Botnet-46 or Scenario 5 in the CTU-13 dataset.](https://mcfp.felk.cvut.cz/publicDatasets/CTU-Malware-Capture-Botnet-46/detailed-bidirectional-flow-labels/capture20110815-2.binetflow)
## Usage
```
python flowvis.py [pca|tsne] [id|tanh|range|std] inputfile.csv
```
we recommend using `tsne` and `tanh` (other methods are used for comparison)
```
python flowvis.py tsne tanh inputfile.csv
```
this would output resulted images and `data.js` which can be viewed using `flowvis.html`
## Results
