David Bezhanyan

Equipped with a robust 13-year background in the banking sector and a progressive specialization in Business Analytics from UC Riverside, my foray into predictive modeling, data mining, and advanced statistical analyses has rendered my skill set increasingly versatile, priming me for success across a variety of data-centric environments.

Predictive modeling of hotel cancellations. Developed R code to clean, subsample, and balance the data using the SMOTE (Synthetic Minority Over-sampling Technique) function, as well as built predictive models (Logistic Regression, Random Forest, KNN) on a comprehensive dataset sourced from a resort hotel with a total of 40,000 observations.

Customer churn prediction in the telecom industry. Addressed missing values, corrected data types and handled outliers. Conducted exploratory data analysis (EDA) using Python libraries such as Matplotlib and Seaborn to visualize data and identify patterns. Created new features, standardized numerical features, and encoded categorical variables to enhance model performance. Split the data into training and testing sets, and built multiple machine learning models, including Logistic Regression, Random Forest, Gradient Boosting and Neural Networks evaluating them using metrics like accuracy and ROC-AUC. Interpreted model outputs to identify key drivers of churn, communicated findings through visualizations, and provided actionable recommendations for reducing churn.