How to Win a Data Science Competition: Learn from Top Kagglers

2020/2021

Academic Year

ENG

Instruction in English

3

ECTS credits

Delivered at:

Department of Applied Mathematics and Informatics (Faculty of Informatics, Mathematics, and Computer Science (HSE Nizhny Novgorod))

Course type:

Compulsory course

When:

2 year, 3 module

Instructor

Razvenskaya, Olga

Course Syllabus

Abstract

The study of this discipline is based on the following courses: • Machine learning • Data analysis methods To master the discipline, students must possess the following knowledge and competencies: • Programming method • Linear algebra Probability and statistics The main provisions of the discipline can be used in their professional activities. https://www.coursera.org/learn/competitive-data-science

Learning Objectives

• The purpose of the discipline is to get acquainted with modern methods of data analysis and ma-chine learning and their use in data analysis competitions

Expected Learning Outcomes

• Be able to choose the method of data processing and perform the data processing by the selected method
• Be able to choose the method of cross validation and evaluate the quality of the selected method of data processing
• Be able to solve the problems of data analysis competitions.

Course Contents

• Data processing
  The tasks of predictive modeling. Preprocessing of data. Collection and processing of data from various sources, texts and images. Advanced feature engineering techniques, generating mean-encodings, using aggregated statistical measures, finding nearest neighbors as a means to im-prove the predictions
• Methodology of cross validation
  Reliable cross validation methodologies. Overfitting and underfitting. Advanced techniques to overcome overfitting and underfitting. Analysis and interpretation of data. Comparisson of data analysis algorithms
• Completions in data analysis
  Examples of competitions. General principals to approach data analysis competition problem. Key issues for a good solution

Assessment Elements

• On-line test
• Examen
• On-line test
• Examen

Interim Assessment

• Interim assessment (3 module)
  0.3 * Examen + 0.7 * On-line test

Bibliography

Recommended Core Bibliography

• Muller, A. C., & Guido, S. (2017). Introduction to machine learning with Python: a guide for data scientists. O’Reilly Media. (HSE access: http://ebookcentral.proquest.com/lib/hselibrary-ebooks/detail.action?docID=4698164)

Recommended Additional Bibliography

• Witten, I. H. et al. Data Mining: Practical machine learning tools and techniques. – Morgan Kaufmann, 2017. – 654 pp.