Web-traffic-anomaly-detection-using-C-LSTM-neural-networks

KO | EN

gitlite — search

#python #hacktoberfest #rust #deep-learning #linux #docker #nodejs #java #awesome #cli #awesome-list #pytorch #mcp #openai

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks

(★ 26)

This project aims to detect the anomalies in Web-Traffic using a C-LSTM architecture.

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Web-traffic-anomaly-detection-using-C-LSTM-neural-networks
This project aims to detect anomalies in Web-Traffic using a C-LSTM architecture.

---

### Web traffic anomaly detection using C-LSTM neural networks

[Web traffic anomaly detection using C-LSTM neural networks ](http://sclab.yonsei.ac.kr/publications/Papers/IJ/2018_ESWA_TYK.pdf)

- This repository contains code to implement the research paper which closely simulate those run by Tae-Young kim in the paper linked above.

- This yahoo s5 dataset has been borrowed from this link https://github.com/harris0704/nbaData16-17/tree/master/Yahoo_S5_Data

---

The model was tested on test data(Which is 30% of total data). The final test results for training the model for 500 epochs(as mentioned in the paper) are:

- Accuracy - 92.7%

- Precision - 58.4%

- Recall - 71.1%

- F1-Score - 64.2%

## Conclusions from the experiment

- The Accuracy and F1-Score of the model increases as we increased the training epochs to 500.

- Changing loss function from mean-square-error to binary-cross-entropy also increased performance of the model because, this is a classification problem binary-cross-entropy loss function gives more accurate results.

- Combining CNN with LSTM greatly helped in detecting patterns and classifying anamolies with good accuracy and precision than most of LSTM networks.

# Instructions to run:

*It is assumed that you have a working copy of Conda installed*

Open up the terminal and type:

git clone https://github.com/sankeerthnalam/Web-traffic-anomaly-detection-using-C-LSTM-neural-networks.git
    cd Web-traffic-anomaly-detection-using-C-LSTM-neural-networks
    conda env create -f environment.yml
    conda activate nnfl_project // optional
    jupyter notebook

Open up 'CLSTM.ipynb' and set path of the dataset directory to import all the files.

You can keep track of training the model through 'accuracy' metrics and if you dont want that just comment the 'Metrics' in model.fit() method.

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks

Web-traffic-anomaly-detection-using-C-LSTM-neural-networks Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README