Магистратура 2025/2026

Рекомендательные системы

Статус: Курс по выбору (Бизнес-аналитика и системы больших данных)

Кто читает: Департамент бизнес-информатики

Где читается: Высшая школа бизнеса

Когда читается: 2-й курс, 1, 2 модуль

Охват аудитории: для всех кампусов НИУ ВШЭ

Преподаватели: Джин Сеунгмин

Язык: английский

Кредиты: 6

Контактные часы: 48

Course Syllabus

Abstract

This course equips Master's students in Business Informactics—particularly non-technical majors—with practical skills to design, evaluate, and deploy recommender systems at a professional level. It provides an intuitive, business-oriented understanding of core algorithms (content-based, collaborative filtering, matrix factorization, and graph-based PageRank), while emphasizing low-code approaches using ChatGPT for rapid prototyping and validation. Key modules include Social Recommender with PageRank for social graph and influence-based recommendations, and Recommender with ChatGPT for conversational interfaces, personalized prompt chains, and explainable AI (XAI) messaging to create "decision-friendly" experiences. By the end, students will master: aligning recommender strategies with business goals, interpreting data and metrics managerially, low-code LLM prototyping, stakeholder communication and governance, and phased roadmaps for resource-constrained environments. Ultimately, the course reframes recommender systems as operational tools driving business outcomes, integrating analytics, product, and strategy competencies. Cf. This course evaluates students with the normalized scores.

Learning Objectives

• Formation of knowledge, skills and development skills of recommender systems for research or industrial purposes.

Expected Learning Outcomes

• Be able to make informed and supported business decisions and recommendations
• Can identify a recommender problem
• Understands the essential methods for recommenders: collaborative filtering, content-based, and matrix factorization
• Able to use embedding for tabular data and recommenders
• Construct variational autoencoders and learn how to sample using this model.
• Be able to construct and estimate recommender systems for content personalization
• able to understand the idea of the latent space for a deep autoencoder-based model and sampling from it
• Build a profile of personal interests
• Build recommendations based on collaborative filtering
• Использует методы prompt инжиниринга и low-code программирования для решения прикладных задач с использованием моделей генеративного ИИ.

Course Contents

• Introduction to Recommender Systems
• Content-Based Filtering
• Collaborative Filtering
• Matrix Factorization and Hybrid Approach
• Deep Learning-Based Recommendation
• Social Recommender with PageRank
• Recommender with ChatGPT
• Capstone Project

Assessment Elements

• Exam
  Test with different types of questions
• Capstone Project
  Student competes their own RecSys model using LLMs
• Attendance
  Students who miss more than 25% of the classes and also fail to complete the presentation assignments may fail this course.
• Capstone Project
  Student competes their own RecSys model using LLMs. This part evaluates if their RecSys is working properly and reproducible.

Interim Assessment

• 2025/2026 2nd module
  The final grade is calculated by weighting and summing the raw scores, followed by min-max normalization, with caps on grade distributions based on student ranks to adjust the overall distribution. The specific formulas and distribution details are as follows. Raw Score Calculation Formula The overall raw score is computed as a weighted sum: Raw Score = 0.2 × Attendance + 0.25 × Exam + 0.25 × Capstone Project + 0.3 × Presentation Min-Max Normalization Formula The raw score is normalized using the minimum (min) and maximum (max) values across all students: Normalized Score = (Raw Score - min(All Raw Scores)) / (max(All Raw Scores) - min(All Raw Scores)) × 100 Grade Distribution and Caps Normalized scores are assigned based on student rank order, with caps applied to maintain the following distribution across all students: 5 points (top tier): Up to 30% of students 4 points: Up to 40% of students 3 points or below: The remaining 30% of students

Authors

• Dzhin Seungmin