People are generating enormous amounts of natural language data in today's digital society, ranging from social media and blogs to forums and product reviews, speeches and news stories. Such natural language text is written by people, and thus can provide insights into their psychology: how they feel about an issue, what kind of people they are, and even how they might behave. This course will introduce students to various methods of text analysis, with a focus on extracting psychological insights. In addition to learning a range of text analysis methods, we will also be discussing the underlying scientific basis that supports various use-cases. The course will be practical and involve substantial coding and implementation as part of homework assignments as well as a semester-long group project. Sample topics include: sentiment analysis, health, online misinformation. Sample Methods include: word-count methods such as sentiment lexicons and the Linguistic Inquiry and Word Count (developed here at UT!); statistical natural language processing tools (e.g., nltk); deep-learning-based Large Language Models (LLMs).
By the end of the course:
We welcome feedback on the course at any point. Feel free to email the instructor directly, or leave anonymous feedback by using the anonymous Google form (URL given in class).
The class group project offers you an opportunity to coalesce all that you have learnt and apply it in a project of your choice. You would have to do a proper literature review, come up with research questions, obtain access to data, apply or modify a method learnt in class, analyze the results, and discuss the inferences.
More information will be provided in class.