200242 SE Theorie und Empirie wissenschaftlichen Arbeitens (Geist und Gehirn) 2 (2024S)

In the course Introduction to Computational Social Sciences, students learn how to use natural language processing tools to analyze the content of large numbers of texts automatically. Students learn to perform sentiment analyses of modern textual sources from the internet, such as movies and books, and online reviews of these media and others, such as video games, music, etc. Furthermore, students learn to analyze historical sources such as fiction works from the past. They can quantify the temporal change of values and preferences expressed in these sources and their relationship to historical events and socio-economic dynamics. Evaluation depends on three short assignments, a final project, and an oral presentation with Q&A. The course roughly covers the following modules:

Introduction to Python. Variables, numbers, strings, lists, loops, modules, and functions;
Introduction to modules. NLTK, Matplotlib, Regular Expressions, etc.;
Bag-of-words analyses. Preprocessing. Finding word synonyms and hyponyms using WordNet;
Word Embeddings. Building semantic clouds and navigating meaning spaces using Word2vec;
Pandas and simple visualizations: Creating datasets, word clouds, and frequency plots. Sentiment analysis. Overview and implementation of lexical and machine learning techniques;
Web scraping. Reading HTML/XML files with Beautiful Soup; Website interaction with Selenium;
APIs. Extracting structured information from social media and popular repositories;
Diachronic Analyses. Using psychometric tools to build valid bags-of-words; historical time series;
Designing studies and testing hypotheses. Relationship between historical events, socioeconomics, and word frequencies. Cross-correlations, lag analyses, and linear mixed models.