Understanding Visitors and their Movements using WiFi Connection Logs

James Pinchin, Matthew Byrne, Iker Perez, Andrew Ward, David Aldred, Sarah Sharples

Abstract:	In this work we examine the concept of using of routinely collected connectivity data to identify the subject interests of a population of visitors to a university open day. We describe and demonstrate the use of a prototype algorithm for performing the subject classification based on the visitors movement and an event timetable. On an open day the university holds events associated with subjects across the university campuses. The nature of the events varies from sample lectures to ‘clinics’ with academic staff. A timetable of events is produced and in combination with a log of the WiFi connections can be used to classify the owner of each device connected to the network. For example a visitor with a device connected to the network for a significant duration in an area being used for a subject specific lecture is assumed to be interested in that subject. This allows the prospective student population to be better understood and open days to be better planned to accommodate visitors. We demonstrate the method using a dataset describing the connections of 936 distinct usernames connected to a network open to visitors on a single open day. Of these distinct usernames 37% were classified to a subject. The number of visitors classified to each faculty was broadly proportional to the size of the faculty.
Published in:	Proceedings of the ION 2017 Pacific PNT Meeting May 1 - 4, 2017 Marriott Waikiki Beach Resort & Spa Honolulu, Hawaii
Pages:	439 - 446
Cite this article:	Pinchin, James, Byrne, Matthew, Perez, Iker, Ward, Andrew, Aldred, David, Sharples, Sarah, "Understanding Visitors and their Movements using WiFi Connection Logs," Proceedings of the ION 2017 Pacific PNT Meeting, Honolulu, Hawaii, May 2017, pp. 439-446. https://doi.org/10.33012/2017.15019
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In