The language that humans use to interact, communicate, and record their thoughts and experiences carries a great deal of information of interest to the health and behavioral sciences. For example, individuals may talk about their symptoms, discuss their emotions, record notes from clinical meetings, or post on social media about their experiences. Unfortunately, this important data often goes unused in scientific analysis because its unstructured nature can make it difficult to model with conventional tools.
Text Mining tools can extract useful information and patterns from unstructured text data in a way that can be easily understood and statistically modeled. This workshop, presented by Timothy Brick, PhD, is intended to provide an introduction to common tools for text mining ranging from simple word counts to the transformer models that underlie ChatGPT, and highlight the ways they can be used to answer questions in the behavioral and health sciences. Topics will include the following:
Finding common topics from a person’s daily reflections
Extracting sentiment and emotions from social media comments
Automatically generating summaries of meeting notes or scientific articles
Creating and modeling text in new ways with deep word embeddings.
The Biostatistics, Epidemiology and Research Design (BERD) Recent Topics in Research Methods seminar series is held during the academic year.
Lectures on fundamental research methods are given in the fall semester, with more advanced topics presented in the spring semester.
Statisticians and methodologists from multiple Penn State departments present these seminars.
7 май 2024
