Study location	Netherlands, Maastricht
Type	Summer Courses, Intensive - Full time
Nominal duration	1 week (2 ECTS)
Study language	English

Tuition fee	€599 one-time

Entry qualification

Enrolled as an Undergraduate student or Undergraduate diploma This course is the first module of a Maastricht Summer School series on Python for (scientific) research. It organically prepares participants for the skills and knowledge taught in the part 2 course. Both modules can be followed independently. In case participants wish to follow only this first part of the series, no prior knowledge of the Python Programming Language is required. Participants would benefit from prior experience in research skills and basis knowledge. The entry qualification documents are accepted in the following languages: English. Often you can get a suitable transcript from your school. If this is not the case, you will need official translations along with verified copies of the original.

Language requirements	English The language of the course is English, so we expect a fluent level and the ability to follow and participate in class.

More information	maastrichtsummerschool.nl

Overview

Course Description
This course takes off where the Maastricht Summer School “Python from Zero to Hero Part 1” ends: turning raw text into meaningful data, or more broadly ‘text mining’.
The goal of text mining is to uncover hidden patterns and/or relationships in large volumes of text data. In Python, text mining is typically done with the help of Machine Learning and Natural Language Processing (NLP) techniques.
In this summer school, participants will be introduced to the ins and outs of text mining with cutting-edge NLP methods. They will learn how to pre-process and wrangle text data so that it becomes machine-readable. Participants will also learn how to harness the capabilities of transformer-based Large Language Models (LLMs) in Python, such as Google’s BERT, for topic modelling and sentiment analysis.

Goals
• Develop an understanding of how computers can learn to understand text.
• Learn the ins and outs of large language models, from basic bag-of-words to more advanced transformer models.
• Learn to pre-process and wrangle text data.
• Provide hands-on experience in implementing text mining techniques (sentiment labelling – topic analysis) using popular Python libraries and frameworks.
• Outreach to generative AI models, augmenting LLMs with your custom data (how to chat with your documents?), and text-to-image models using LangChain and Huggingface models.

Recommended Literature
All course materials (including slides, scripts, audio-visual materials, and written media) and will be made available to all participants.

Teaching Methods
Lectures, Assignments, Research, Skills, Coding

Assessment Methods
There is no formal examination for this course, but attendance and participation are required. Participants who have actively attended and participated will receive a certificate.

Course Coordinator
Dr. Arnoud Wils

Programme structure

Full course description
Full Course Description

Large Language Models Workshop: How computers are learning to understand texts

In this comprehensive four-day workshop, we will delve into the fundamentals of natural language processing (NLP) and explore how computers can be trained to understand text data. We will cover a range of topics, from basic concepts like bag-of-words and n-gram models, to advanced techniques such as vector semantics and embeddings, transformers, and pretrained language models. Participants will also learn about various NLP applications including sentiment and affect labelling, topic analysis, and outreach to retrieval models.

The workshop will provide hands-on experience in implementing these techniques using popular libraries and frameworks, and participants will have the opportunity to work with real-world sample data to apply their newfound knowledge. By the end of the workshop, attendees will have a solid understanding of how large language models work and how they can be used to unlock insights from text data.

Day 1: NLP Fundamentals

• Introduction to NLP and its applications
• Bag-of-words and n-gram models
• Vector semantics and embeddings

Day 2: Advanced NLP Techniques

• Transformers and pretrained language models
• Fine-tuning and masked language models

Day 3: NLP Applications

• Sentiment and affect labelling
• Topic analysis

Day 4: Outreaching to Retrieval Models

• How to chat with your own documents
• Other NLP applications and case studies

This workshop is designed for anyone with a good foundation in Python and who is interested in learning about large language models and their applications in NLP. Join us for an exciting journey into the world of large language models and discover how they can help you unlock the power of text data.