Title: Threefold similarity analysis: a case study on crowdsourcing feeds

Authors: Kaixu Liu; Gianmario Motta; Tianyi Ma; Ke Fan

Addresses: Department of Industrial and Information Engineering, University of Pavia, Pavia, Italy ' Department of Industrial and Information Engineering, University of Pavia, Pavia, Italy ' Department of Industrial and Information Engineering, University of Pavia, Pavia, Italy ' Department of Industrial and Information Engineering, University of Pavia, Pavia, Italy

Abstract: Crowdsourcing is a valuable social sensing for the smarter city. We present a framework of crowdsourcing feeds similarity analysis from a threefold point of view, namely image, text, and geography, which is based on similarity analysis, founded on a sequence that goes from coarse to thinner similarity filters. The main idea is to extract feeds within a specific geographic range, and then to analyse similarity of image colour and text in clustered feed sets. The framework enables to identify feeds that report the same issue, and hence to filter redundant information. Based on proved methods and algorithms, such framework has been implemented in a software application, called CITY FEED, which is used by the Municipality of Pavia.

Keywords: crowdsourcing; smart city; image similarity analysis; text similarity analysis; clusters; text semantic analysis.

DOI: 10.1504/IJITM.2019.099807

International Journal of Information Technology and Management, 2019 Vol.18 No.2/3, pp.327 - 345

Received: 24 Jul 2015
Accepted: 11 May 2016

Published online: 10 May 2019 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article