|
Title:
|
ON THE SUITABILITY OF TIME SERIES ANALYSIS ON DATA WAREHOUSES |
|
Author(s):
|
Jose Zubcoff , Juan Trujillo , Alfredo Cuzzocrea |
|
ISBN:
|
978-972-8924-40-9 |
|
Editors:
|
Jörg Roth, Jairo Gutiérrez and Ajith P. Abraham (series editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Time series analysis, data warehousing, knowledge discovery, multidimensional modeling, conceptual modeling. |
|
Type:
|
Full Paper |
|
First Page:
|
17 |
|
Last Page:
|
24 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
Nowadays, Time Series (TS) analysis is done more as an art rather than a science. It is frequently applied to flat files,
obtained from external data sources such as operational databases. The main reason of this isolated data mining
perspective is the lack of a methodology that integrates the conceptual modeling of the repositories, the pre-processing,
and the posterior phases of the Knowledge Discovery in Databases (KDD) process. Such an isolated perspective of KDD
can lead us to duplicate time-consuming tasks such as those of pre-processing steps. On the other hand, considering the
main goals from the beginning of the knowledge discovery process allows us to improve data quality significantly.
Previous work accomplishes the multidimensional model design of the Data Warehouse (DW). This work provides a
conceptual model for the TS analysis on DW via establishing an abstract view on common features of this mining
technique by reusing existing models of the underlying DW, thus achieving a comprehensive methodology allowing us to
represent, map, and execute TS analysis tasks on DW. To show the suitability of our approach in a practical
experimentation, we have mapped the conceptual models defined for a case study on platform specific models given by a
commercial database system. As a result, hiding low-level details allows us to present the entire KDD process to users in
terms of familiar concepts, thus facilitating the communication among human resources involved in the process. |
|
|
|
|
|
|
