Link Search Menu Expand Document
UC Berkeley, Fall 2020

Developing mindset metrics for business and societal applications

Partner: Ming Hsu, Haas School of Business, Academic

Overview

Project Description

Recent advances in artificial intelligence and machine learning have made it possible to uncover subtle patterns and hidden trends in large-scale real-world data. They offer researchers and practitioners powerful tools to efficiently derive novel insights and predictions that are otherwise expensive or even impossible to obtain. Among the most exciting of these recent advances is reverse-engineering the human mind and, in parallel, engineering more humanlike machine learning systems, which has begun to help us address some of the most challenging questions about the nature and origins of human thought, knowledge, and biases. This research project aims to develop and apply cutting-edge data science, especially natural language processing, techniques for challenging problems about human mind and behavior. Student researchers will be exposed to cutting-edge research at the intersection of natural language processing and computational social sciences, with the unique opportunity of validating data-driven insights with empirical data of human behavior. Students will explore application and extensions of these models in one of two projects: (i) quantifying and linking stereotypes in news and social media to real world measures of prejudice and discrimination, and (ii) developing metrics of consumer ““mindset”” capturing how consumers think, feel, and behave toward different products and brands.
Specifically, student researchers will be trained to

  1. mine online content ranging from news media, social media, and social networks
  2. develop computational models that utilize semantic relationship embodied by real-world text corpora to capture cognitive processes
  3. combine these models with computational social neuroscience models to predict real-world and laboratory behavior and outcomes

Qualifications: Strong computational and programming skills, interest in computational social science. Desirable, but not essential: Experience with natural language processing and text analysis Weekly Hours: to be negotiated”

Expected Deliverable

Written report, web app, posting dataset.

What would a successful semester look like to you?

Student learning in NLP data analysis, behavioral modeling, and linking text data with behaviors and outcomes.

Data

Models

Conclusion