Computing the value of a crowd

  • Authors:

  • Manas S. Hardas;Lisa Purvis

  • Affiliations:

  • Spigit, Inc., Pleasanton, CA, USA and Computer Science Department, Kent State University, Kent, OH;Spigit, Inc., Pleasanton, CA

  • Venue:
  • SBP'12 Proceedings of the 5th international conference on Social Computing, Behavioral-Cultural Modeling and Prediction
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes that the "value" of a crowd can be defined in terms of the overall engagement of the individuals within the crowd and that engagement is a function of certain characteristics of the crowds such as small world-ness, sparsity and connectedness. Engagement is hypothesized as messages being exchanged over the complex network which represents the crowd and the "value" is calculated from the entropy of message probability distributions. An initial random network is passed through a process of entropy maximization and the values of some structural properties are recorded with the changing topology to study the corresponding behavior. We show that as the small world-ness and connectedness of a crowd increases and the sparsity decreases, the engagement in the crowd increases.