Search

In the paper, we first thoroughly examine and prove properties of the participation index of spatio-temporal sequential patterns. Then, we introduce notions of a closure of a spatio-temporal sequential pattern and a closed spatio-temporal sequential pattern, as well as investigate and prove their properties. In particular, we prove that the set of all participation index strong closed spatio-temporal sequential patterns constitues a lossless representation of all participation index strong spatio-temporal sequential patterns. We also propose an algorithm, called CST-SPMiner, for discovering all participation index strong closed spatio-temporal sequential patterns. CST-SPMiner is an adaptation of the STBFM algorithm, which was proposed recently for the discovery of spatio-temporal sequential patterns with high participation index. While STBFM uses the CSP-tree structure for time-efficient candidate patterns generation and evaluation, CST-SPMiner uses it also for fast identification of closed patterns. Efficiency and effectiveness of our algorithm were verified on real crime data for Boston. [ABSTRACT FROM AUTHOR]

Spatial co-location pattern mining is a sub field of data mining which is used to discover interesting patterns which are expressed as co-location rules. The objects that are frequently located in certain region are expressed as spatial co-locations. It presents a challenge for finding co-location patterns as the traditional data is considered discrete whereas the spatial objects are embedded in a continuous space. For this a join-less approach is proposed, but as the data size increases, a large amount of computation time is devoted to find co-location rules as the approach is purely sequential. We propose a parallelized join-less approach which finds the spatial neighbor relationship in order to identify co-location instances and co-location rules. The proposed work decreases the computation time drastically as it uses a Map-Reduce framework. This paper presents precise and completeness of the new approach. Finally, an experimental evaluations using synthetic data sets show the algorithm is computationally more efficient. [ABSTRACT FROM AUTHOR]

Abstract: This paper develops a composite participation index (PI) to identify patterns of transport disadvantage in space and time. It is operationalised using 157 weekly activity–travel diaries data collected from three case study areas in rural Northern Ireland. A review of activity space and travel behaviour research found that six dimensional indicators of activity spaces were typically used including the number of unique locations visited, distance travelled, area of activity spaces, frequency of activity participation, types of activity participated in, and duration of participation in order to identify transport disadvantage. A combined measure using six individual indices were developed based on the six dimensional indicators of activity spaces, by taking into account the relativity of the measures for weekdays, weekends, and for a week. Factor analyses were conducted to derive weights of these indices to form the PI measure. Multivariate analysis using general linear models of the different indicators/indices identified new patterns of transport disadvantage. The research found that: indicator based measures and index based measures are complement each other; interactions between different factors generated new patterns of transport disadvantage; and that these patterns vary in space and time. The analysis also indicates that the transport needs of different disadvantaged groups are varied. [Copyright &y& Elsevier]