Mining Semantic Web data using Correspondence Analysis - ExampleSet

This workflow describes how to learn from the Semantic Web's data using a data transformation algorithm 'Correspondence Analysis'.

The input to the workflow is a feature vector developed from a RDF resource. The loaded example set is divided into training and test parts. These sub-example sets are used by the Correspondence Analysis operators (encapsulate the Correspondence Analysis data transformation technique) which processes each feature at a time and transform the data into a different space. This transformed data is more meaningful and helps the learner to improve classfication peformance. The tranformed data and the calculated distance matrices can be observed using Distance Matrix operator, which gives a better understanding of underlying process.

Please visit the following URL to download the released plugin together with code and relevant documentation:

code.google.com/p/rapidminer-semweb/

Looking forward for the feedback