Author(s) |
Detterer, Dion
Kwan, Paul H
|
Publication Date |
2012
|
Abstract |
Traditional Chinese Medicine (TCM) relies heavily on interactions between herbs within prescribed formulae. However, given the combinatorial explosion due to the vast number of herbs available for treatment, the study of herb-herb interactions by pure human analysis is impractical, with computer aided analysis computationally expensive. Thus feature selection is crucial as a pre-processing step prior to herb-herb interaction analysis. In accord with this goal, a new feature selection algorithm known as a Co-evolving Memetic Wrapper (COW) is proposed: COW takes advantage of recent developments in genetic algorithms (GAs) and meme tic algorithms (MAs). evolving appropriate feature subsets for a given domain. As part of preliminary research. COW is demonstrated to he effective in selecting herbs in the TCM insomnia dataset. Finally, possible future applications of COW are examined, both within TCM research and in broader data mining contexts.
|
Citation |
New Frontiers in Applied Data Mining, p. 361-371
|
ISBN |
9783642283208
9783642283192
|
Link | |
Publisher |
Springer
|
Series |
Lecture Notes in Artificial Intelligence
|
Edition |
1
|
Title |
COW: A Co-evolving Memetic Wrapper for Herb-Herb Interaction Analysis in TCM Informatics
|
Type of document |
Book Chapter
|
Entity Type |
Publication
|
Name | Size | format | Description | Link |
---|