Title: Crowd estimation at a social event using call data records

Authors: V.P. Sumathi; Kittusamy Kousalya; V. Vanitha; J. Cynthia

Addresses: Department of Computer Science and Engineering, Kumaraguru College of Technology, Coimbatore, Tamilnadu, India ' Department of Computer Science and Engineering, Kongu Engineering College, Perundurai, Tamilnadu, India ' Department of Computer Science and Engineering, Kumaraguru College of Technology, Coimbatore, Tamilnadu, India ' Department of Information Technology, Kumaraguru College of Technology, Coimbatore, Tamilnadu, India

Abstract: Segregating attendees of the event from other regular visitor, paves way for new applications such as measuring event success and outdoor advertising. Traditional methods of crowd estimation at points of interests such as shopping malls, cinema halls, stadium and exhibitions using sample analysis of visitor through participant count obtained by number of tickets sold. In unstructured (free-to-view) event, the above said method does not help in accurate crowd estimation and social analysis. The objective of this paper is using call data records to estimate the crowd at an unstructured event in a city. The study was conducted at Indian Institute of Science (IISc) campus, Bengaluru on a special event day. To avoid breach of privacy, anonymised call data records have been used. The crowd estimated using call data records shows positive correlation with ground truth-value.

Keywords: social analysis; crowd estimation; call data records; CDRs; unstructured event; free-to-view event; attendance estimation.

DOI: 10.1504/IJBIS.2018.091866

International Journal of Business Information Systems, 2018 Vol.28 No.2, pp.246 - 261

Received: 02 Aug 2016
Accepted: 14 Nov 2016
Published online: 20 May 2018 *

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