Feature reduction improves classification accuracy in healthcare

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Hamid R. Nemati, Professor (Creator)
Fereidoon "Fred" Sadri, Professor (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site: http://library.uncg.edu/

Abstract: Our work focuses on inductive transfer learning, a setting in which one assumes that both source and target tasks share the same features and label spaces. We demonstrate that transfer learning can be successfully used for feature reduction and hence for more efficient classification performance. Further, our experiments show that this approach increases the precision of the classification task as well.

Additional Information

IDEAS 2018: Proceedings of the 22nd International Database Engineering & Applications Symposium, June 2018. Pages 193–198
Language: English
Date: 2018
classification, transfer learning, feature reduction, classification accuracy

Email this document to