Classification of URLs as Malicious or Benign
This Repo consists of the following notebooks:
- Web Metrics Scrapper : Scrapes dataset for the project
- Data and Features : Additional Feature generation and their explaination
- TestDataPrep : Preparation of test dataset for model perforamnce evaluation
- Model Fitting : Model Fitting on Training Data and performance evaluation on the test data
Libraires used:
- sci-kit learn
pip install sklearn
- keras
pip install keras
There are the following models in this repo:
- LR.pickle : Logistic Regression Classfier
- RFC.pickle : Random Forest Classifier
- Scaler.pickle : Standard Scaler for preprocessing required for Neural Network
- model.h5 : Artificial Neural Network Classfier