An evaluation of combinations of lossy compression and change-detection approaches for time-series data

详细信息    查看全文

• 作者：Gregor Hollmig ^{gregor.hollmig@student.kit.edu} ; Matthias Horne ^{matthias.horne@student.kit.edu} ; Simon Leimkü ; hler ^{simon.leimkuehler@student.kit.edu} ; Frederik Schö ; ll ^{schoell@ira.uka.de} ; Carsten Strunk ^{carsten.strunk@student.kit.edu} ; Adrian Englhardt ^{adrian.englhardt@student.kit.edu} ; Pavel Efros ; ^{pavel.efros@kit.edu} ; Erik Buchmann ^{erik.buchmann@kit.edu} ; Klemens Bö ; hm ^{klemens.boehm@kit.edu}
• 关键词：Time series ; Compression ; Change detection
• 刊名：Information Systems
• 出版年：2017
• 出版时间：April 2017
• 年：2017
• 卷：65
• 期：Complete
• 页码：65-77
• 全文大小：1372 K
• 卷排序：65

文摘

We describe applications scenarios for our framework. We explain the design decisions behind our comparison study. We present our selection of compression, change-detection and optimization techniques. The evaluation shows that result quality highly depends on the dataset. We present several heuristics regarding change-detection algorithms.