Mining spatiotemporal association rules from spatiotemporal databases between two different fixed dates. Issue 2 (2015)

• Record Type: Journal Article
• Authors: Alouaoui, Hana; Turki, Sami Yassine; Faiz, Sami
• Is Part Of: International journal of knowledge engineering and data mining. Volume 3:Issue 2(2015)
• Page Start: 190
• Page End: 207
• Publication Date: 2015
• Subjects: spatiotemporal databases -- data mining -- spatiotemporal relationships -- association rules mining -- fixed dates -- hidden relationships -- pattern extraction -- frequent itemsets -- Tunisia -- Tunis; Knowledge representation (Information theory) -- Periodicals; Data mining -- Periodicals; 006.305
• Languages: English
• ISSNs: 1755-2087

Abstract

Mining frequent itemsets from large spatiotemporal databases is a hard task due to the existence of hidden spatiotemporal relationships. The aim of our proposal is to look for spatiotemporal association rules that relate properties of reference objects like large towns with properties of other spatial task-relevant objects. The extracted patterns are different relationships relating the spatial objects during time periods. We propose a four-step approach; the core step is to extract the spatiotemporal relationships. The second and third steps are devoted respectively to frequent spatiotemporal itemsets generation and spatiotemporal association rules extraction. The fourth step is the refinement of the extracted rules. To prove the applicability of our method, we conduct experimentation on a spatiotemporal database describing the city of Tunis in the dates 1987 and 2001. On the basis of our proposed measures namely the spatial closeness relevance and the time subsequence, only interesting spatiotemporal association rules are retained.