Toward mining of spatiotemporal maximal frequent patterns

POPELÍNSKÝ, Lubomír and Jan BLAŤÁK. Toward mining of spatiotemporal maximal frequent patterns. In Proceedings of ECML/PKDD Workshop on Mining Spatio-Temporal Data (MSTD),. Porto: UP, 2005, p. 31-40.

Basic information

• Original name: Toward mining of spatiotemporal maximal frequent patterns
• Name in Czech: K dolování v prostorově-časových datech
• Authors: POPELÍNSKÝ, Lubomír (203 Czech Republic, guarantor) and Jan BLAŤÁK (203 Czech Republic).
• Edition: Porto, Proceedings of ECML/PKDD Workshop on Mining Spatio-Temporal Data (MSTD), p. 31-40, 10 pp. 2005.
• Publisher: UP

Other information

• Original language: English
• Type of outcome: Proceedings paper
• Field of Study: 10201 Computer sciences, information science, bioinformatics
• Country of publisher: Portugal
• Confidentiality degree: is not subject to a state or trade secret
• RIV identification code: RIV/00216224:14330/05:00014243
• Organization unit: Faculty of Informatics
• Keywords in English: data mining; spatiotemporal data
• Tags: data mining, spatiotemporal data

Abstract

We show that propositional spatiotemporal logic PSTL is a powerful tool for mining in various spatiotemporal data including environmental and medical data, keystroke dynamics data or text. We introduce a refinement operator for a fragment of $PSTL$, $ST_0$ and %, and present frequent patterns mined with RAP. describe the ILP system GRAPE for mining first-order frequent patterns in spatiotemporal data. We also show that in the classification task %the use of this refinement operator can %decrease computational cost and that the use of frequent patterns as new features result in an accuracy increase.

Abstract (in Czech)

We show that propositional spatiotemporal logic PSTL is a powerful tool for mining in various spatiotemporal data including environmental and medical data, keystroke dynamics data or text. We introduce a refinement operator for a fragment of $PSTL$, $ST_0$ and %, and present frequent patterns mined with RAP. describe the ILP system GRAPE for mining first-order frequent patterns in spatiotemporal data. We also show that in the classification taskthe use of frequent patterns as new features result in an accuracy increase.

Links

• MSM0021622418, plan (intention): Name: DYNAMICKÁ GEOVIZUALIZACE V KRIZOVÉM MANAGEMENTU

Investor: Ministry of Education, Youth and Sports of the CR, Dynamic Geovisualisation in Crises Management