In this python project, I am trying to build a Classification Machine Learning model to predict banknotes are genuine or forged.
In real world this could mainly be any of the followings.
- Fraud detection
- counterfeit detection
- quality control
- authentication of banknotes
There are several valuable Business Impacts and Pottential Benefits which we can define here.
- Reducing finantial losses
- Improve customer trust
- Enhancing operational efficiency
- or meeting any regulatory requirements
For this project stackholders possibly be
- any Banking system
- finantial institutions
- law enforcement agencies
- or any regulatory bodies
This project is based on Bank Notes Authentication UCI dataset dataset. I'm using the Kaggle's version of it.
I will be using
- Machine Learning Algorithms for classification banknotes.
- Various Python libraries to visualize different insights along the way
Descriptive Statistics wil be used to derive valueable insights from the data.
Following Machine Learning algoritms will be evaluated and select the best performing model as the final model.
- Logistic Regression
- Random Forest
- KNN Classifire
- Support Vector Machine Classifire
I have used pyforest library bundle for this project.
!pip install pyforest
- Hyperparameter tuning using GridSearchCV()
- Foundation Methodology for Machine Learning Project