Newest technological developments enable fast collection of large datasets, opening new challenges and opportunities for applications in many areas, including insurance. This course focuses on the problem of knowledge discovery from structured and unstructured insurance data, including large datasets (big data). The course gives a comprehensive introduction into the procedures and quantitative methods for knowledge discovery from data and connects two integral aspects vital to data-driven decision making: applied statistics and machine learning. Visualization and exploratory data analysis procedures are introduced, together with the basic principles of statistical inference. Classification, regression, clustering and dimensionality reduction problems are considered. A special part of the course is directed at a selection of machine learning models, with an introduction to deep learning. In addition to familiarizing with modern knowledge discovery concepts, attention is given to the application of this knowledge through working with real world data. The course will combine lectures, guest lecturers, case study analyses and team work.
Postgraduate spec. study
Trevor Hastie, Robert Tibshirani, Jerome Friedman (2013.), The Elements of Statistical Learning, Springer Science & Business Media
Roxy Peck, Chris Olsen, Jay L. Devore (2011.), Introduction to Statistics and Data Analysis, Cengage Learning
Ronald E. Walpole, Raymond H. Myers, Sharon L. Myers, Keying Ye (2016.), Probability and Statistics for Engineers and Scientists,
L3 English Level