230011 Processing social media data (S) (SoSe 2022)

Social Media (such as Twitter, Facebook, Reddit, YouTube, and so on) is a big part of our daily lives. This big unstructured data is a very valuable source for researchers to understand individual and societal tendencies. To make use of such data and transform it into manageable and scientific or application-oriented research topics, the field of computational linguistics provides many tools which we will explore throughout this semester with this course.

After a general introduction and a hands-on primer on programming in Python (e.g., with NumPy, sci-kit-learn), we will dive into several exciting topics, including scraping data from social media and text preprocessing, data exploration, and text classification.

The course will be taught in English.

In this seminar, students learn to

• break down a research question/problem into manageable components
• develop an analytical approach to address a research problem in computational sociolinguistics.
• crawl a linguistically valuable social media data
• experiment with data from popular social media platforms
• differentiate various types of social media data and methods to process/analyze them
• apply different classical machine learning or deep learning algorithms in Python Environment to get insights for specific research questions,
• interpret the result of data analysis,

To successfully pass, we ask participants
• to hand in 2-3 small homework assignments
• to present a paper or Python library/package
• to submit a 1-2 page research question/hypothesis summary