Title: RFID-based solution for galleries and museums visit modelling using Markov model, BBN's and MAP decisions

Authors: Petar Solic, Nikola Rozic, Josko Radic

Addresses: Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Department of Communication and Information Technologies, University of Split, Rudjera Boskovica b.b. 21000 Split, Croatia. ' Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Department of Communication and Information Technologies, University of Split, Rudjera Boskovica b.b. 21000 Split, Croatia. ' Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Department of Communication and Information Technologies, University of Split, Rudjera Boskovica b.b. 21000 Split, Croatia

Abstract: RFID technologies are becoming increasingly popular and widely used in many applications. Tags can be used for environment and habit monitoring, healthcare applications, home automation and pedestrian or vehicle traffic control. This paper describes the method of building a robust N-state Markov model that describes visitor|s behaviour in a gallery room. The built model can be used in planning of exhibitions, in modelling of visitor|s preferences, and/or in generation of predictions related with exhibition lasting, expected sales and pricing. Presented system performance improvements are realised through Bayesian belief network (BBN) and maximum a posterior probability (MAP) decision approximation algorithm.

Keywords: algorithms; visitor behaviour; Markov modelling; visitor modelling; RFID based applications; visitor tracking; Bayesian networks; MAP decision making; gallery visitors; museum visitors; museums; galleries; exhibition planning; visitor preferences; radio frequency identification.

DOI: 10.1504/IJIIDS.2010.036893

International Journal of Intelligent Information and Database Systems, 2010 Vol.4 No.6, pp.532 - 551

Published online: 15 Nov 2010 *

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