Text Mining: Transforming Text into Knowledge

Course goals

Overview

With the rapid growth of digital textual data in many areas of science, there is a growing need for automated tools that can analyse, classify, and interpret this type of data. Text mining techniques can be used to create a structured representation of text, making its content more accessible to users and researchers. Text mining applications are everywhere: social media, web search, advertising, email, customer service, healthcare, marketing, etc. During the course, students will actively learn how to apply text mining methods to data analysis and how to use them together with natural language processing and machine learning techniques on real data problems. The course has a strong practical focus: students will gain hands-on experience in Python by applying the methods to real data during the course and interpreting the results.

Learning Goals

The aim of this course is to provide students with an understanding of the principles, problems, techniques, and solutions associated with text mining and to enable them to gain knowledge of how recent advances in text mining relate to innovative approaches to organising, characterizing, finding and exploiting large amounts of textual information in the search for new knowledge. On completion of the course, students should be able to:

1. Explain and use text preprocessing and representation techniques.
2. Describe a text analysis system and its components, both optional and mandatory.
3. Define a text mining pipeline given a practical data science problem.
4. Implement all steps of a text mining pipeline: feature extraction, model learning, model evaluation.
5. Analyse and reflect on the different techniques used in text mining, the parameters required, and the problem solved.
6. Understand and apply some of the state-of-the-art methods in text mining and natural language processing.
7. Plan and carry out a text analysis experiment.

Relation between Assessment and Objective

1. The final exam:
   1. Demonstrate theoretical knowledge of the concepts underlying text mining and text analysis (Learning Goals 1, 2, 3)
   2. Apply the above concepts to hypothetical research scenarios (Learning Goals 2, 4, 5)
   3. Interpret text analysis experiments and their output (Learning Goals 6, 7)
2. The assignments:
   1. Analyze, implement and work with text datasets (Learning Goals 2, 3)
   2. Implement a text mining pipeline for real text analysis situations using reproducible reporting procedures (Learning Goals 3-5)
   3. Understand and apply state-of-the-art methods in text mining (Learning Goals 3-7)

Content

In this course, we will introduce a variety of basic principles, techniques, and modern advances in text mining. Topics to be covered include basic natural language processing techniques, text representation, text classification, feature selection, text clustering and topic models, word embedding, deep neural networks and introduction to (large) language models.

Note: Students who cannot comply with the entrance requirements mentioned below will be asked to provide further information on their eligibility. The course coordinator will decide on their eligibility.

Assumed knowledge
Basic knowledge/motivation in programming and data science.