Exploratory Data Analysis (EDA) of Bank Marketing Campaign

Project Overview

This project aims to analyze the impact of marketing campaigns on customers’ decisions to open a term deposit account. By utilizing exploratory data analysis (EDA) and statistical methods, the study identifies key patterns, relationships, and actionable insights for optimizing marketing strategies.

Objectives

1. Perform Data Profiling and Exploratory Data Analysis (EDA) using pivot tables and visualizations.
2. Conduct statistical analyses, including:
   • Correlation Analysis
   • Chi-Square Test
   • Linear Regression
3. Provide key insights and recommendations to improve campaign effectiveness.

Dataset

• Source: Kaggle – Bank Marketing Dataset
• Size: 45,211 rows and 17 columns
• Key Variables:
  • campaign: Number of contacts made during this campaign.
  • y: Target variable indicating if the term deposit was subscribed (yes/no).
  • balance, duration, pdays, job, education, marital status, etc.

Key Findings

1. Statistical Insights

Chi-Square Test

• Significant Relationship: The number of campaigns has a statistically significant impact on term deposit subscription.
• Direction: Negative relationship – more campaigns reduce the likelihood of success.

Linear Regression

• Coefficient: -0.0076 → Each additional campaign reduces the likelihood of success by 0.76%.
• R-squared: 0.005 → Campaigns explain only 0.5% of the variation, indicating other factors influence success.

2. Recommendations

• Optimize Campaigns: Focus on the quality, not quantity, of interactions.
• Avoid Over-Contacting: Limit contacts to prevent diminishing returns.
• Segment Clients: Target responsive clients based on features such as balance and education.

Methods Used

1. Exploratory Data Analysis (EDA)

• Univariate Analysis: Histograms, boxplots, and frequency distributions.
• Multivariate Analysis: Heatmaps and scatter plots.
• Pivot Tables: Analyzing correlations between categorical variables.

2. Statistical Analysis

• Correlation Analysis: Pearson and Spearman correlations.
• Chi-Square Test: Examining independence between categorical features.
• Linear Regression: Evaluating trends and impact of campaign contacts on deposit subscription.

Project Structure

• data/: Contains raw and preprocessed dataset files.
• notebooks/: Jupyter notebooks for EDA and statistical analysis.
• scripts/: Python scripts for data processing and visualization.
• README.md: Project documentation.

How to Run the Project

1. Clone the repository:

   git clone https://github.com/hijirdella/Bank-Campaign-EDA.git
   

2. Install required dependencies:

   pip install -r requirements.txt
   

3. Run the Jupyter notebook:

   jupyter notebook notebooks/EDA_and_Analysis.ipynb
   

Visualizations

• Campaign frequency distribution
• Balance vs. term deposit subscription
• Correlation heatmap of numerical features

Contact

• Author: Hijir Della Wirasti
• GitHub: Bank Campaign EDA Repository
• LinkedIn: Hijir Della Wirasti
• Email: hijirdw@gmail.com

This project provides in-depth insights into the effectiveness of bank marketing campaigns and suggests improvements to maximize conversion rates. Future work can include predictive modeling to enhance targeting strategies.