A range of data sources across the internet, such as google search terms, twitter topics and Facebook messages, amongst others, can be viewed as kinds of sensors from which information might be extractable about trends in the expression of matters of concern to people. We focus on the problem of how to identify emerging trends after the original textual data has been processed into a quantitative form suitable for the application of machine learning techniques.We present some preliminary ideas, including an agent-based implementation and some early results, about the application of artificial prediction markets to such data, taking the specific domain of syndromic surveillance (early stage recognition of epidemics) as an example, using publicly available data sets.
|Number of pages||14|
|Journal||CEUR Workshop Proceedings|
|Publication status||Published - 9 Jan 2014|
|Event||Sintelnet WG5 Workshop on Crowd Intelligence: Foundations, Methods, and Practices, CROWD 2014 - Barcelona, Catalonia, Spain|
Duration: 8 Jan 2014 → 9 Jan 2014
ASJC Scopus subject areas
- Computer Science(all)