Activity recognition from accelerometer data is a major research area within mobile and wearable computing [1,2]. The goal of the thesis is to (i) collect accelerometer (and potentially other sensor modalities) measurements from customers during their regular shopping activities and to (ii) evaluate the feasibility of using accelerometer data for recognizing key events related to purchasing behaviour of customers within a retail environment (e.g., inspecting items, putting items in a shopping basket or cart, taking items out etc.).

References:

[1] Lane, N. D.; Miluzzo, E.; Lu, H.; Peebles, D.; Choudhury, T. & Campbell, A. T. A Survey of Mobile Phone Sensing IEEE Communications Magazine, 2010, 48, 140 -150

[2] Atallah, L. & Yang, G.-Z. The use of pervasive sensing for behaviour profiling -- a survey Pervasive and Mobile Computing, 2009, 5, 447-464