Research question
Which customer and campaign features predict a successful marketing outcome?
Predictive modeling · Classification
Bank Marketing
A predictive modeling project focused on identifying the customer characteristics and campaign conditions associated with successful bank marketing outcomes.
Outcome
Classify whether a customer subscribes to the marketed bank product
Decision use
Prioritize likely responders and improve campaign resource allocation
Research question and scope
This project evaluates whether customer and campaign characteristics can predict a successful response to a bank marketing campaign.
The target is a binary subscription outcome. The model is intended to support prioritization, not replace campaign judgment or customer-level review.
Method
The workflow includes data preparation, exploratory analysis, feature encoding, and a deep-learning classification model.
Model performance is evaluated with a held-out test set and classification measures that distinguish correct targeting from missed opportunities and false positives.
Model interpretation
Customer profile, prior interaction history, and campaign contact conditions provide useful signals for predicting response.
The strongest practical use is ranking prospects for outreach, with performance monitored against the cost of additional contacts.
Primary visual
Model performance and interpretation
Downloads
Project files and datasets
Access the completed modeling report and final presentation.