A Mathematical Approach to Identifying and Forecasting Shifts in the Mood of Social Media Users

Abstract

Social media offers a promising opportunity to identify and understand the direction of the mood of people using these platforms, just as conventional radars help one identify and understand physical motion. To date, many of the methods used to analyze social media in this way are qualitative, relying on the inputs of human subject matter experts. Those quantitative approaches which have been validated are in their infancy. This paper presents a new quantitative approach to characterizing the mood of social media users that can complement existing qualitative methods. This novel method combines a validated computer program (LIWC) with a mathematical algorithm to follow trends in past and present moods and detect breakpoints where those trends changed abruptly. First steps have also been taken to further develop this method so that it can also predict future trends in mood and possibly forecast related events. Validation is an important aspect of this part of the overall study. Finally, preliminary guidance for putting the output of the breakpoint analysis and forecasting into context is provided. The paper concludes with an overview of directions for continued research.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2012
Accession Number
AD1108576

Entities

People

  • Les Servi
  • Sara B. Elson

Organizations

  • MITRE Corporation

Tags

Communities of Interest

  • Biomedical
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Computer Programs
  • Computers
  • Corporations
  • Data Set
  • Data Sets
  • Delphi Method
  • Dictionaries
  • Digital Data
  • Dynamics
  • Human Behavior
  • Indicators
  • Language
  • Media
  • Military Applications
  • Online Communications
  • Psychology
  • Situational Awareness
  • Social Environment
  • Social Media
  • Social Networking Services
  • Social Psychology

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computational Modeling and Simulation
  • Economics