I combine academic theories and practical approaches in the data science area. I have a PhD degree in physics and DSc in AI. My current interest lies in the area of Machine Learning & Deep Learning, Predictive Analytics, Pattern Recognition, Time Series Analytics, Natural Language Processing, Computer Vision, Social Network Analytics, Business Intelligence, Anomaly Detection, Reinforcement Learning, Numeric Modeling, Risk Assessment, Reliability Theory, Financial Modeling, Supply Chain Analytics, Customer’s Behavior Analytics, Demand Prediction, Dynamic Price Optimization. In predictive analytics models, I combine machine learning and Bayesian inference that is an effective approach for forecasting and risk assessment in business processes with non-Gaussian statistics.
I work on state-of-the-art predictive analytics solutions, take part in Kaggle competitions where I have a Master degree and 3 gold medals for top positions in leaderboards. As a teammate, I won one Kaggle competition (“Grupo Bimbo Inventory Demand”).
I concentrated my main efforts on developing predictive models, features engineering, combining predictive models into diversified stacking ensembles, combining parametric models with algorithmic machine learning models and probabilistic models, estimating prediction uncertainty and risk assessments.
My Links:
Resume: https://bit.ly/3KqE7Wf
DSc thesis “Methods of intellectual analysis of consolidated data for decision-making support”: https://bit.ly/3poKrmO
arXiv e-prints: https://bit.ly/3nufuhk
LinkedIn Articles: https://www.linkedin.com/in/bpavlyshenko/detail/recent-activity/posts/
Kaggle profile: https://www.kaggle.com/bpavlyshenko
Kaggle competition 1st place solution: https://bit.ly/3qEAiox