The Finance Project involves analyzing customer data from a bank to predict the likelihood of a customer accepting a personal loan offer. The dataset contains information on customer demographics, their relationship with the bank, and their response to previous loan campaigns. The goal is to build a predictive model using supervised learning techniques to identify customers who are likely to accept a personal loan offer, aiding the bank's marketing efforts.
Thera Bank aims to convert its liability customers into personal loan customers while retaining them as depositors. With a successful conversion rate in a previous campaign, the bank's retail marketing department seeks to improve target marketing effectiveness with minimal budget.
The primary objective is to predict the likelihood of a liability customer accepting a personal loan offer. This involves building a model using supervised learning methods such as Logistic Regression, K-Nearest Neighbors (KNN), and Naive Bayes Algorithm.
- Data Exploration: Understanding the dataset's features, distributions, and correlations.
- Data Preprocessing: Handling missing values, fixing errors, and scaling attributes.
- Model Building: Utilizing Logistic Regression, KNN, and Naive Bayes classifiers.
- Model Evaluation: Assessing model performance through accuracy metrics and confusion matrices.
- Optimization: Iteratively refining models by considering feature inclusion, scaling, and algorithm selection.
- Logistic Regression achieved an accuracy of 95% with the 'Experience' feature included.
- KNN yielded an accuracy of 96% without the 'Experience' feature.
- Naive Bayes demonstrated comparable performance with and without the 'Experience' feature, achieving an accuracy of approximately 85%.