Course is confirmed
Machine Learning (ML) and Artificial Intelligence (AI) are increasingly integrated into our daily lives. From self-driving cars to voice-activated home assistants and the effects that enhance videos on platforms like TikTok, ML applications are all around us.
This course begins with a refresher on R and RStudio before introducing the core types of ML and their appropriate applications. We will cover supervised and unsupervised learning techniques, including classification and regression methods (e.g., random forests, regression modeling, gradient boosting, and deep learning). You will also explore dimension reduction techniques such as autoencoders and UMAP, which simplify complex datasets by reducing the number of dimensions, and clustering methods like K-means, hierarchical clustering, and DBSCAN for grouping similar observations.
In addition to structured data from surveys and experiments, you will learn how ML techniques can process and analyze unstructured data, such as images and text. This includes applying neural networks for image classification, topic modeling with LDA, and deep learning methods like BERT for text classification.
The course focuses on understanding when to use specific ML techniques and gaining hands-on experience applying them in R/RStudio. It is designed to bridge the gap between theoretical knowledge and practical implementation in the context of social sciences.
Learning objectives
- Identify and evaluate suitable Machine Learning (ML) techniques for various research problems in social science.
- Apply a range of ML techniques, including classification, regression, clustering, and dimension reduction, to structured and unstructured datasets.
- Develop hands-on experience in implementing ML methods using R and RStudio.
- Understand the distinctions between supervised and unsupervised learning approaches and their applications.
- Analyze and interpret the results of ML models to draw meaningful conclusions for research purposes.
