Paper Info
Title
Effective data mining: a data warehouse-backboned architecture
Abstract
An effective Data Mining (DM) system for mining multiple-level knowledge from Data Warehouse (DW), DB and flat files of raw data is proposed. The DW represents the backbone of the proposed architecture. Intermediate, as well as final results of mining are incorporated into the DW for efficient processing of further queries. A Markov Chain mathematical model is developed for managing data dependency and consistency in the DW. An adaptive hybrid view technique is introduced to manage storage space. DM and OLAP technologies are closely integrated. The mining and OLAP kernel includes generic analysis modules for performing a wide spectrum of applications. Active data mining is adopted to support knowledge-driven business processes. Continuously gathered business data is partitioned according to application-dependent time periods. Active mining uses these partitioned data sets to discover rules and key business indicators for each time period.
Year
Venue
Keywords
1998
CASCON
partitioned data,time period,data warehouse-backboned architecture,key business indicator,active mining,data dependency,business data,knowledge-driven business process,raw data,mining multiple-level knowledge,active data mining,effective data mining,business process,data mining,data warehouse,spectrum
Field
DocType
Citations 
Data warehouse,Data mining,Data stream mining,Data dependency,Business process,Computer science,Raw data,Flat file database,Business intelligence,Online analytical processing,Database
Conference
3
PageRank 
References 
Authors
0.46
11
3
Name
Order
Citations
PageRank
Khalil M. Ahmed1192.38
Nagwa M. El-Makky26311.48
Yousry Taha3201.74