Бакалавриат
2026/2027
Научно-исследовательский семинар: Введение в машинное обучение
Статус:
Курс обязательный (Политология и мировая политика)
Где читается:
Санкт-Петербургская школа социальных наук
Когда читается:
4-й курс, 1 модуль
Охват аудитории:
для всех кампусов НИУ ВШЭ
Преподаватели:
Аркатов Дмитрий Александрович
Язык:
английский
Кредиты:
2
Контактные часы:
14
Course Syllabus
Abstract
The research seminar is devoted to data analysis technologies. It presents the reserach activity from the perspective of quantitative technoques.
Learning Objectives
- To introduce political science students to the core concepts and practical tools of machine learning, with a focus on developing intuition for predictive modeling and hands-on experience applying methods in R.
- To enable students to critically evaluate the use of machine learning in political and social research, balancing methodological understanding with awareness of interpretability, fairness, and ethical considerations.
Expected Learning Outcomes
- Differentiates between explanatory and predictive approaches in political science research, and explain how supervised and unsupervised machine learning methods address different types of questions.
- Applies basic machine learning techniques in R — including regression with regularization, classification with trees and random forests, and clustering/PCA — to real political and social science datasets.
- Evaluate model performance and interpret results using appropriate metrics (e.g., RMSE, accuracy, ROC, cluster separation, explained variance), and critically reflect on the tradeoffs between interpretability and predictive power.
- Assess the opportunities and limitations of machine learning in political science, including issues of fairness, bias, and ethical implications in applying algorithms to social and political data.
Course Contents
- Introduction to Machine Learning
- Classification and Decision Trees
- Clustering
- Fairness and Interpretability
Assessment Elements
- Final ProjectThe final project is the capstone of the course, allowing students to integrate the concepts and methods learned into an applied research-style exercise. Students must choose a dataset (from an approved list or one of their own, subject to instructor approval) and apply at least one supervised method (e.g., regression, logistic regression, decision tree, random forest) and one unsupervised method (e.g., clustering, PCA). The project deliverable is a 5–7 page paper accompanied by R code as an appendix. The paper should include: (1) a clear research question framed in political science terms, (2) a description of the chosen data and preprocessing steps, (3) implementation of machine learning methods with evaluation of model performance, (4) substantive interpretation of results, and (5) reflection on ethical or fairness considerations relevant to the context. The project assesses not only technical implementation but also the student’s ability to communicate findings in a way that connects machine learning methods to political science inquiry.
- Paper-Based TestsThese are brief, 15-minutes in-class assignments designed to evaluate your retention of core concepts and terminology from previous lectures and practices. No electronic devices are allowed and the tests are closed-book.
- HomeworksHomeworks consist of short, hands-on coding assignments designed to reinforce the methods introduced in class and seminars. Each homework requires students to apply one or more machine learning techniques to a provided political science or social dataset. Students must submit their R scripts along with output (tables, plots, or model evaluation metrics) and a brief interpretation of the results in plain language, explaining what the model reveals about the data. The focus is on practical implementation, correct use of methods, and meaningful interpretation, rather than extended conceptual discussion.
- ExamThe exam is designed to evaluate your knowledge in full. This is a written assignment consisting of two comprehensive questions. These questions are randomly drawn from the pool of 20 provided to you in advance. This ensures your understanding of the material without worrying about the unexpected topics.
Interim Assessment
- 2026/2027 1st module0.3 * Paper-Based Tests + 0.201 * Exam + 0.2 * Homeworks + 0.299 * Final Project
Bibliography
Recommended Core Bibliography
- 9780262046824 - Kevin P. Murphy - Probabilistic Machine Learning - 2022 - MIT Press - https://search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=2932689 - nlebk - 2932689
- Matt Wiley, & Joshua F. Wiley. (2019). Advanced R Statistical Programming and Data Models : Analysis, Machine Learning, and Visualization. Apress.
- Miroslav Kubat. (2017). An Introduction to Machine Learning (Vol. 2nd ed. 2017). Springer.
- Miroslav Kubat. An Introduction to Machine Learning. Springer, 2015 (296 pages) ISBN: 9783319200095: — Текст электронны // ЭБС books24x7 — https://library.books24x7.com/toc.aspx?bookid=117295
- Rogers, S., & Girolami, M. (2016). A First Course in Machine Learning (Vol. 2nd ed). Milton: Chapman and Hall/CRC. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=1399490
Recommended Additional Bibliography
- Machine learning fundamentals : a concise introduction, Jiang, H., 2021