Бакалавриат
2025/2026
Обработка естественного языка
Статус:
Курс по выбору (Прикладной анализ данных)
Где читается:
Факультет компьютерных наук
Когда читается:
4-й курс, 1, 2 модуль
Охват аудитории:
для своего кампуса
Преподаватели:
Пахуров Фёдор Константинович
Язык:
английский
Кредиты:
5
Контактные часы:
56
Course Syllabus
Abstract
The course "Natural Language Processing" is dedicated to the introduction to natural language processing (NLP) problems at the intersection of disciplines such as machine learning, deep learning, and linguistics. The course consists of two parts: (1) introduction to classical NLP, (2) advanced NLP methods, including state-of-the-art models and Large Language Models (LLM) applications.
Learning Objectives
- Study basic tasks and methods of natural language processing and text analysis
- Study modern neural network models for natural language processing
- Acquire knowledge of software systems and tools for text processing and analysis
- Study LLM and its applications to the practical tasks
Expected Learning Outcomes
- Be able to apply basic word processing and analysis techniques
- Be able to formulate and solve problems related to language modeling and specialized problems on text data
- Know the ethical aspects of word processing
- Be able to use and adapt LLM for applied NLP tasks
Course Contents
- NLP Introduction. Statistical text analysis.
- Vector text representation models
- Texts classification
- Language models
- Sequence to sequence tasks
- Attention, Transformer
- Common NLP tasks
- Pretrained language models. Transfer learning
- Modern pretrained models
- Large Language Models (LLM)
- LLM Training
- Efficient fine-tuning & prompting
- LLM & Modern NLP tasks; LLM Evaluation, Safety and fairness
- RAG and long-context models
- Advanced LLM usage - tools, agents, MCP
- Model speed-up and compression; uncertainty estimation, fairness
