Power Laws & Language: Can We Predict Population-Level Decision Making?

Mark Keane
School of Computer Science & Informatics, 
University College Dublin

The last few years have seen an explosion of research that predicts human 
behaviour using language, specifically, words that are available online.  
Cultural changes can be tracked from GoogleBooks, search queries can be 
used to predict flu epidemics and tweets frequencies can predict
movies box-office takings.  Most of this research suggests that what 
people write down, is a fairly good proxy for what they think and, 
subsequently, decide to do; especially, if those words are
in a very large data set. 

This talk presents one further example of this research trend, based on 
work done with Aaron Gerow (TCD) on predicting stock market movements 
from changes in the distributional patterns of language. 
Gerow & Keane (2011, IJCAI-11) show that power laws of words in 
financial articles (from the New York TImes, BBC and Financial Times) 
undergo systematic changes as one enters a bubble market.  The
wider implications of this work for population-level language analyses 
and characterising human behaviour from words are explored.


Short Biography

Since 1998, Prof. Mark Keane has been Chair of Computer Science at University
College Dublin. From 2004-2007 he was Director of ICT (2004-2006) and Director
General (2006-2007) at Science Foundation Ireland (SFI) where he oversaw a
700M+ euro research investment. He advised the Irish Government on its 3.7B
euro Strategy for Science, Technology & Innovation (SSTI). He was also VP of
Innovation & Partnerships at UCD (2007-2009). He has a BA (UCD) and PhD (TCD)
in Cognitive Psychology and previously worked in University of London, the Open
University, Cardiff University and Trinity College Dublin (FTCD, 1994). Prof.
Keane has published 150+ articles, has a H index of 34 reflecting almost 6000
Google Scholar citations, including 12 publications with >100 citations.  He
currently carries out research on analogy, metaphor, the semantics of motion,
surprise and big data language analyses.