Εξόρυξη δεδομένων σε πολυδιάστατες χρονοσειρές
Mining multi-dimensional timeseries
View/
Keywords
Data mining ; Εξόρυξη δεδομένωνAbstract
In our day, time series data is reproduced, updated and stored daily in various databases. The time series data collected, in addition to the fact that they are increasing day by day, tend to take on a multidimensional form. In this way, multidimensional time series are created that require new and better ways of data mining, indexing and searching. In this work we describe existing techniques of the iSAX family, used for one-dimensional time series data mining operations. For multidimensional time series data mining operations, we describe and use an existing method, which is an extension of the previous techniques, called hyperSAX. To demonstrate its usefulness, the hyperSAX technique is applied to two-dimensional time series for indexing and similarity search operations and the time it takes to build the index is measured.
Postgraduate Studies Programme
Εφαρμοσμένη ΣτατιστικήDepartment
Σχολή Χρηματοοικονομικής και Στατιστικής. Τμήμα Στατιστικής και Ασφαλιστικής ΕπιστήμηςNumber of pages
52Language
GreekCollections
Except where otherwise noted, this item's license is described as
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές
Attribution-NonCommercial-NoDerivatives 4.0 Διεθνές