Understanding the Past, Present, and Future Economic Behaviors of Customers for Financial Products Over Time

View Webinar Recording

You must be located within the United States to access this dataset

 

DATA

AI and Analytics for Business is delighted to announce an exciting new dataset from one of the largest Information Service Companies in the US. The project sponsor is actively seeking new ways to leverage their data for matching consumers to credit services. This unique dataset captures the financial activity of over 2 million consumers and contains the opening and closing of several types of accounts, collections histories, bankruptcy filings, and changes in credit attributes and credit scores over time.

While the sponsor is open to any potential use of the data, they are particularly interested in proposals that explore:

• Are there general-purpose approaches to detecting anomalous consumer behavior that could be used to detect fraud and reporting errors?
• Bayesian modeling approaches for risk score analysis
• Is it possible to completely automate the process of translating information from raw credit files into attributes/scores without human intervention?

You must be located within the United States to access this dataset

 

RESEARCHER DATA ACQUISITION AND MODELING ENVIRONMENT

This is the first AIAB Research Opportunity to pilot an entirely new user experience!
The data sponsor has built a clustered Hadoop data environment, and will offer the following features to awarded teams:

• A unique on-boarding process
• Individually customized access to a virtual desktop infrastructure
• A host of tools to explore data and perform analytics
• An online question/request system and live help desk

To learn more about the data and business context, interested faculty and doctoral students can attend watch an archived webinar recording (registration required). This webinar will offer a live demo of data access and the research environment and executives from the project sponsor will be available for Q&A. The webinar will also be archived for those who can’t attend live.

View Webinar Recording