%0 Generic
%A Sari Aslam, N
%A Cheng, T
%A Cheshire, J
%A Zhang, Y
%D 2019
%F discovery:10127905
%I GISRUK
%K Trip purposes, smart card data, COP-KMEANS, semi-supervised clustering
%T Trip purpose identification using pairwise constraints based semi-supervised clustering
%U https://discovery.ucl.ac.uk/id/eprint/10127905/
%X Clustering of smart card data captured by automated fare collection (AFC) systems has traditionally  been viewed as an unsupervised method. However,some additional information about human behaviour  is available in addition to the smart card data points that can facilitate better partitioning of the data. In  this paper, such prior knowledge is translated into pairwise constraints and used with the COPKMEANS clustering algorithm to identify user activities. The effectiveness of the method was  evaluated using performance evaluation measures by comparison of the results with the ground truth.  The results demonstrate that pairwise constraints significantly enhance the accuracy of the clusters.
%Z This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions.