PatZip: Pattern-Preserved Spatial Data Compression

This paper presents a compression method, PatZip, to improve the efficiency of spatial pattern mining methods. PatZip can avoid overcompression and stop automatically before pattern is destroyed. Compared with existing compression methods, PatZip is deterministic and its result is reproducible, and...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Qian, Yu, Zhang, Kang, Huynh, D. T.
Format:	Tagungsbericht
Sprache:	eng
Schlagworte:	Applied sciences Artificial intelligence Compression Method Compression Stage Computer science control theory systems Data processing. List processing. Character string processing Exact sciences and technology Learning and adaptive systems Memory organisation. Data processing Neighbor Graph Pattern Mining Representative Point Software
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	This paper presents a compression method, PatZip, to improve the efficiency of spatial pattern mining methods. PatZip can avoid overcompression and stop automatically before pattern is destroyed. Compared with existing compression methods, PatZip is deterministic and its result is reproducible, and original data can be easily recovered. The compression process is data-driven and parameter-free, and requires only O(nlogn) time for n data points.
ISSN:	0302-9743 1611-3349
DOI:	10.1007/11430919_84