Mindset Segmentation Prediction for Affluent Population
Customers at MassMutual are defined within 5-segment attitudinal segmentation framework. In order to run marketing campaign on affluent prospective customers, business stakeholders asked Data Science to predict mindset segments for prospects within certain age, income, and net worth criteria. Third party vendor data provided the mindset segmentation framework, then this data was matched to out prospect database. The models were learned from matched data and applied to unmatched data. To address class imbalance issues, SMOTE tool was applied. The best model was based on Random Forest algorithm via the sklearn package.
Anna Arakelyan is a Lead Data Scientist in the Sales and Marketing domain, Assistant Vice President at MassMutual. She holds a PhD in Economics from CUNY Graduate Center and an M.S. and B.S. in Quantitative Economics from Lomonosov Moscow State University. In her doctoral studies she focused on applied econometric analysis of social networks. Her areas of expertise are econometrics, statistics and probability, machine learning, causal inference, and design of experiments.