Anomaly Detection In Temporal Data Mining - Softcover

Onat, Mehmet Yavuz

Softcover

ISBN 10: 3659797499 ISBN 13: 9783659797491

Publisher: LAP LAMBERT Academic Publishing, 2015

Synopsis

Temporal data mining is a title for data mining techniques executed over temporal data. The major goals of temporal data mining are; indexing, clustering, classification, prediction, summarization, anomaly detection and segmentation. In temporal data, anomaly detection or novelty detection is the identification of interesting patterns. Several anomaly detection algorithms have been proposed in the literature. However, there are limited number of studies that compare these methods. In this study, Heuristically Ordered Time series using Symbolic Aggregate Approximation (HOT-SAX), Pattern Anomaly Value (PAV), Wavelet and Augmented Trie (WAT) and Multi-Scale Abnormal Pattern Detection Algorithm (MPAV) anomaly detection methods were compared by using synthetic and real temporal data sets. Also, temporal data representation techniques were compared in terms of anomaly detection. R statistical programming language was used for analysis.

About this title

About the Author

Birth: 02.09.1988, Ankara (Turkey). Bachelor and Master's Degree: Statistics, Dokuz Eylül University, İzmir (Turkey). Temporal data mining is a growing field. I would like to continue my study during my PhD and introduce a brand-new algorithm or representation technique to the field.

Edition details

• Publisher: LAP LAMBERT Academic Publishing
• Publication date: 2015
• Language: English
• ISBN 10: 3659797499
• ISBN 13: 9783659797491
• Binding: Paperback
• Number of pages: 72