Wang, Ye, Lee, Kyungmi, and Lee, Ickjai (2016) Visual analytical tool for higher order k-means clustering for trajectory data mining. In: Lecture Notes in Computer Science (9992) pp. 507-518. From: AI 2016: 29th Australasian Joint Conference on Artificial Intelligence, 5-8 December 2016, Hobart, TAS, Australia.

PDF (Published Version) - Published Version Restricted to Repository staff only

Abstract

Trajectories are useful sources to understand moving objects and locations. Many trajectory data mining techniques have been researched in the past decade. Higher order information providing suggestions to what-if analysis when the best possible option is not feasible is of importance in dynamic and complex spatial environments. Despite of the importance of higher order information in trajectory data mining, it has received little attention in literature. This paper introduces new visualisation methods for determination of higher order k-means clustering for trajectory data mining. This paper proposes a radar chart-like visualisation for geometrical and directional higher order information and a k-means clustering technique for trajectory higher order information. This paper also demonstrates the usefulness of proposed visualisation methods and clustering technique with a case study using real world datasets.

Item ID:	47543
Item Type:	Conference Item (Research - E1)
ISBN:	978-3-319-50127-7
Keywords:	higher order information, visual analytics, spatio-temporal data mining, trajectory data mining
Date Deposited:	15 Mar 2017 00:24
FoR Codes:	46 INFORMATION AND COMPUTING SCIENCES > 4605 Data management and data science > 460502 Data mining and knowledge discovery @ 100%
SEO Codes:	89 INFORMATION AND COMMUNICATION SERVICES > 8902 Computer Software and Services > 890202 Application Tools and System Utilities @ 100%
Downloads:	Total: 4
	More Statistics