Бакалавриат
2025/2026
Машинное обучение для бизнеса
Статус:
Курс обязательный (Международная программа по бизнесу и экономике)
Кто читает:
Департамент экономики
Где читается:
Санкт-Петербургская школа экономики и менеджмента
Когда читается:
4-й курс, 1 модуль
Охват аудитории:
для своего кампуса
Язык:
английский
Кредиты:
4
Контактные часы:
28
Course Syllabus
Abstract
This modular course is an introduction to machine learning tools. The average course completion time may vary depending on the student's initial training. The prerequisite for mastering the course is knowledge of mathematics at the secondary education level and the basics of the Python programming language. Students' academic success is assessed through programming assignments in the form of contests, as well as written controls in the form of tests. The final exam is a contest with a presentation of the solutions proposed by the students. With the help of course assignments, basic methods of data preprocessing are worked out. model construction, interpretation of results. The course does not involve lectures, all theoretical materials are provided to students in practical classes.
Learning Objectives
- • Understanding the basic rules of syntax, data types, and built-in constructs • Create custom preprocessing pipelines • Mastering the main Python machine learning library: sklearn • Formation of basic skills in using Python as a classification and forecasting tool
Expected Learning Outcomes
- the student is able to explain the main types of data and the formulation of the research task
- the student is able to create basic machine learning models for regression and classification tasks
- the student is able to find and eliminate syntactic and logical errors in scripts
- the student is able to analyze the results obtained in order to describe economic processes
Course Contents
- 1. Problem statement, data, pipelines, metrics
- 2. Linear regression models
- 3. Classification: logistic regression, imbalance
- 4. Trees and ensembles for tabular data
- 5. Interpretation, stability, drift
- 6. Time series for economists
- 7. Automatic
Assessment Elements
- Test
- Contest 1Contest 1 is a competition on the Kaggle site for students of the course. The students' task is to get the correct answers to the prediction tasks uploaded to the website. To get the correct answers to the prediction problem, the student must build a machine learning model by applying knowledge from the course topics. The number of attempts by the student is not limited until the end of the time of the control element. Students are prohibited from sharing solutions. After uploading, the student's result is reflected on the leaderboard.
- Contest 2Contest 2 is a competition on the Kaggle site for students of the course. The students' task is to get the correct answers to the prediction tasks uploaded to the website. To get the correct answers to the prediction problem, the student must build a machine learning model by applying knowledge from the course topics. The number of attempts by the student is not limited until the end of the time of the control element. Students are prohibited from sharing solutions. After uploading the solution, the student's result is reflected on the leaderboard.
- Contest 3Contest 3 is a competition on the Kaggle site for students of the course. The students' task is to get the correct answers to the prediction tasks uploaded to the website. To get the correct answers to the prediction problem, the student must build a machine learning model by applying knowledge from the course topics. The number of attempts by the student is not limited until the end of the time of the control element. Students are prohibited from sharing solutions. After uploading the solution, the student's result is reflected on the leaderboard.
- Exam contestThe Exam contest is a competition on the Kaggle site for students of the course. The students' task is to get the correct answers to the prediction tasks uploaded to the website. To get the correct answers to the prediction problem, the student must build a machine learning model by applying knowledge from the course topics. The number of attempts by the student is not limited until the end of the time of the control element. Students are prohibited from sharing solutions. After uploading the solution, the student's result is reflected on the leaderboard. Students submit solutions in .ipynb format for verification and prepare a presentation justifying the chosen methods for the solution. Students can take work in groups of up to 4 people.
Interim Assessment
- 2025/2026 1st module0.15 * Contest 1 + 0.15 * Contest 2 + 0.15 * Contest 3 + 0.4 * Exam contest + 0.15 * Test
Bibliography
Recommended Core Bibliography
- 9781800206571 - Serg Masís - Interpretable Machine Learning with Python : Learn to Build Interpretable High-performance Models with Hands-on Real-world Examples - 2021 - Packt Publishing - https://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=2901980 - nlebk - 2901980
Recommended Additional Bibliography
- Alpaydin, E. (2014). Introduction to Machine Learning (Vol. Third edition). Cambridge, MA: The MIT Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=836612