This project develops a machine learning model to classify customers based on their features and predict whether they will make a deposit into their newly opened account, using a dataset from a Portuguese bank's marketing campaign.
- Data Mining Course - Spring 2024
The dataset represents a marketing campaign by a Portuguese bank, containing customer information. The dataset contains 11162 samples, with 16 features. The features include:
- Demographic information: age, job, marital status, education
- Financial information: default, housing, loan
- Marketing campaign information: contact, month, day of week, duration, campaign, pdays, previous, poutcome
- Target variable: deposit (binary categorical)
The EDA notebook provides an exploratory data analysis of the dataset, including:
- Data cleaning and preprocessing
- Visualization of the data using various plots and charts
The classification notebook develops a machine learning model to classify customers based on their observed features. The model is trained using a variety of algorithms, including KNN and decision trees. The performance of each model is evaluated using metrics such as accuracy, precision, and recall.
The clustering notebook applies clustering algorithms to the dataset to identify patterns and group similar customers together. The algorithms used include k-means and hierarchical clustering. The results of the clustering analysis are visualized using various plots and charts.