Home // ICDS 2016, The Tenth International Conference on Digital Society and eGovernments // View article
Authors:
Lorenzo Gabrielli
Davide Guido
Fosca Giannotti
Luca Bastiani
Keywords: Open Data; eDemocracy; eGovernment; Citizen-Government eModels, Data Mining, Administrative elections
Abstract:
Can we predict the outcome of future elections? Many politicians wonder if they will be re-elected and often they rely on polls conducted on a small sample of the population. In this work, we propose a data-driven approach that, given the past expenditure of city mayors, considers what are the most important aspects that determine the re-election. Our empirical results show the emergence of a particular expenditure threshold: mayors who invest in current and capital expenditure over such threshold during the mandate are more likely to be re-elected, while those who invest differently are more likely not to be re-elected. The impact of this research is to provide a new analytical tool that objectively shows to a public administrator if his actions will lead to the re-election.
Pages: 18 to 21
Copyright: Copyright (c) IARIA, 2016
Publication date: April 24, 2016
Published in: conference
ISSN: 2308-3956
ISBN: 978-1-61208-521-0
Location: Venice, Italy
Dates: from April 24, 2016 to April 28, 2016