Bohdan combines academic theories and practical approaches in the data science area. He is Data Scientist at SoftServe company and an associate professor (Ph.D.) in Electronics and Computer Technologies Faculty at Lviv National University and has more than 70 scientific publications. His current scientific interest lies in the area of quantitative linguistics, machine learning, predictive analytics, computer vision, social networks mining, business intelligence, time series analytics, numeric modelling, risk assessment, reliability theory, financial modelling. He has strong practical experience in retail and supply chain analytics, customers behavior analytics, fraud detection. In predictive analytics models, he combines machine learning and Bayesian inference that is an effective approach for forecasting and risk assessment in the business processes with non-Gaussian statistics. He works on the state of the art predictive analytics solutions, taking part in Kaggle competitions where he has Master degree and 3 gold medals for top positions in leaderboards. As a teammate, he won one Kaggle competition among nearly two thousand teams where his team proposed the best solution for sales forecasting in the chain of nearly 800 thousand stores.
The main points of presentation: Different approaches for sales time series predictions. Risk assessments in business analytics. Deep learning approaches for business time series predictive analytics. Using probabilistic models in business intelligence. Regularized linear models for time series regression models. Using multilevel stacking and bagging models in business time series prediction. Combining linear, machine learning and Bayesian models in predictive analytics.