Polina Ananyevskaya

Medical data research: correlation analysis, classification problems for small samples with categorical variables and missing values (R, Python,Statistica)

Data quality insights (missing values, broken frequencies, gaps) and recomendations for quality improving provided. Lags in data collections estimated and lag adjusting automated schema created. Automated anomalies detection mechanism created for time series. Seasonal adjustment for different types of seasonality implemented. The prototype created for portfolio stress testing.

Data targeting research: dealing with big data (hadoop, hiveQL); user segmentation according to their profiles, interests and conversion type using machine learning methods; text classification; data quality control (R, Java, Python for running hive queries, Tableau for reporting)

Projects completed: features importance for sales prediction, rejected contract prediction based on continuous and discrete historical data (R)