Meeting Summary: Large-scale text from online sources and naturalistic conversations can provide rich insights into the content and operation of human minds. Yet the methods to mine these texts, including methods of natural language processing (NLP) and machine learning, have only recently been introduced in the social sciences and in the study of social cognition. Additionally, the theoretical perspectives and toolkits offered by the psychological study of language have rarely infiltrated the techniques of computer scientists and methodologists. For this working group, we are bringing together a small group of interdisciplinary scientists from psychology and computer science, to learn from each other and map out a program of collaborative research that can motivate progress on questions of joint interest that currently operate in parallel with no connecting lines.
The two-day working group will focused on two broad themes: (1) understanding the machine learning approach of word embeddings and how it may be applied to questions of psychological interest; and (2) understanding the diverse approaches used by psychologists, including topic modeling and dictionary methods (i.e., LIWC) and how they may inform ongoing work on NLP in computer science.