Bank Marketing
Marketing data set used to predict whether a customer will take out a term deposit, based on more than 40,000 prospecting telephone calls with socio-economic variables.
4 CSV files including 1 full version with 41,188 rows and 20 columns
Apache 2.0
Description
The dataset Bank Marketing compiles data from direct marketing campaigns conducted by a Portuguese bank. Each line represents a phone call made to promote a term deposit, with information such as age, profession, marital status, banking history, etc. The objective is to predict whether the customer will accept the offer or not.
What is this dataset for?
- Train supervised classification models (e.g. decision tree, logistic regression, random forest)
- Optimize marketing campaigns according to the customer profiles most likely to subscribe
- Analyze the influence of socio-demographic variables on banking behavior
Can it be enriched or improved?
Yes, you can enrich the dataset by adding behavioral indicators based on real or simulated data (e.g. browsing history, marketing emails). It is also possible to recode certain variables to make them more usable (binning, encoding, etc.) or to train models on reduced versions for benchmarks.
🔎 In summary
🧠 Recommended for
- Marketing analysts
- Machine learning students
- Customer scoring demonstrations
🔧 Compatible tools
- Scikit-learn
- LightGBM
- Orange
- Jupyter
- Table
💡 Tip
Test different combinations of variables to detect the customer profiles that are most sensitive to phone calls.
Frequently Asked Questions
Does this dataset allow customer segmentation?
Yes, it is possible to group customers by behavior, age, profession, or past responses to create targeted marketing segments.
What is the target variable?
The target variable is “y”, which indicates whether the customer accepted (“yes”) or refused (“no”) the proposed product.
Is the dataset adapted to a predictive scoring project in business?
Yes, it is an excellent test base for experimenting or prototyping scoring systems in a banking environment.
