- Research Name Clickstream data analysis
- Client Quicken Loans
- Category PRACTICUM PROJECTS
- RESEARCH YEAR 2019
The Quicken Loans Family of companies has a lot of website traffic that is client-facing. QL is looking to utilize the data to understand client behavior through their digital activities.
The Quicken Loans team gave the Wayne State team two goals to achieve over the course of the project.
- Characterize frequent website navigation patterns of users
- Learn navigation patterns of users likely to purchase a QL product
- Develop mechanisms for timely intervention and conversion management
The team developed algorithms based on deep-dive statistical analysis and constructive data visualizations, applied Association rule mining, Random forest, XG-Boost algorithm to predict navigation pattern based on historical data and identified, measured and recommended improvement strategies, resulted in improving the mortgage lending process.
Sumanth Raj Tumiki, Christopher Harrison, Dhanashree Dandare, and Xia Liu