[ML] Credit Assessment
Develop and optimize classification models to assess individuals' creditworthiness based on economic profiles using the German credit dataset.
Objective
Determine creditworthiness of individuals based on their economic profiles.
Utilize credit assessments to decide eligibility for financial products.
Dataset Overview
Source: UCI Machine Learning Repository
Description: Profiles of German individuals with creditworthiness labels (1 or 0).
Total Records: 1,000
Number of Features: 20
Key Analysis Steps
Data Preparation
Handle duplicates and missing values.
Encode categorical data and transform features.
Perform exploratory data analysis (EDA) for insights.
Model Development
Build and evaluate classification models, including Random Forest Classifier.
Optimize models using random and grid search methods.
Feature Selection and Analysis
Perform statistical tests to select key features.
Interpret results using Feature Importance and Shapley Values.
Key Learning Outcomes
Techniques for analyzing mixed data types (categorical and numeric).
Feature selection methods using statistical testing.
Advanced model optimization techniques and interpretability using FI and Shapley Values.