Title :
Web services composition for distributed data mining
Author :
Ali, Ali Shaikh ; Rana, Omer F. ; Taylor, Ian J.
Author_Institution :
Sch. of Comput. Sci., Cardiff Univ., UK
Abstract :
A Web services-based toolkit for supporting distributed data mining is presented. A workflow engine is provided within the toolkit to enable a user to compose Web services to implement particular point solutions. Three types of Web services are provided to implement data mining functions: (1) classifiers; (2) clustering algorithms; and (3) association rules. Additional capability is made available through GNUPlot and Mathematica to enable visualisation of the output. Data sets may be read from the local filespace, or streamed from a remote location (provided the algorithm being used has support for streaming). A study is presented to illustrate the use of the toolkit.
Keywords :
Internet; data mining; software tools; workflow management software; GNUPlot; Mathematica; Web services; association rules; clustering algorithm; distributed data mining; local filespace; output visualization; remote location streaming; toolkit; workflow engine; Algorithm design and analysis; Breast cancer; Classification algorithms; Clustering algorithms; Data analysis; Data mining; Data visualization; Machine learning algorithms; Pipelines; Web services;
Conference_Titel :
Parallel Processing, 2005. ICPP 2005 Workshops. International Conference Workshops on
Print_ISBN :
0-7695-2381-1
DOI :
10.1109/ICPPW.2005.87
