Course 2022-2023 a.y.

1.       Introduction to machine learning in economics and social sciences

2.       Natural Language Processing

3.       Causal Machine Learning

4.       Applications of supervised machine learning

• Measuring media bias 

5.       Applications of unsupervised machine learning

• Measuring Ideology

6.       Applications of topic models

• Measuring central bank communication

7.       Applications of word embeddings

• Measuring biases in text

8.       Applications of image classification

• Measuring characteristics of images

·         Explain the how machine learning can be used to answer new questions in the social sciences

·         Describe different machine learning methods and their application

·         Recognize potential applications of machine learning in real-world data

·         Illustrate the advantages and disadvantages of different machine learning methods in various contexts

·         Analyze the results of state-of-the-art economic research

·         Evaluate the validity of machine learning application

·         Assess how sensitive results are to the choice of machine learning algorithm

·         Critically weigh different machine learning algorithms against each other

·         Discuss how machine learning is broadening the fields of social science research.

·         Hypothesize about future applications of machine learning 

• Face-to-face lectures
• Exercises (exercises, database, software etc.)
• Group assignments

The learning experience of this course includes face-to-face lectures by the instructors. The instructor will additionally provide some coding examples and exercises. In addition, students are also asked to prepare one group presentation on a research paper. These presentations are used to introduce students to the critical evaluation of empirical research and the contribution of papers to the scientific literature. Further, the presentations are intended to stimulate discussions about potential machine learning applications to social science questions. This will allow the students to develop their own ideas about future machine learning projects.

The course assessment takes place on a “pass or fail” bases. The student presentation will be used to assess the above-mentioned learning outcomes. The student presentation will ask students to deeply involve themselves with state-of-the-art economic research that uses machine learning and provide a presentation to their fellow students. In this way the presentation aims to assess if students are able:

• To apply the knowledge and skills from the class to a new research paper
• To discuss the validity of chosen machine learning approaches
• To describe how machine learning helps to answer the relevant research question.
• To articulate shortcomings of the chosen machine learning strategies
• To suggest future applications of similar machine learning strategies

For non-attending students the assessment is based on a written report they will be asked to write about one research paper.

Teaching materials are announced before the start of the course and indicated or uploaded to the Bboard platform.