A demo of Gaussian-based Decision Tree, which is designed for classification on small dataset, e.g., there are only two sample for training in each category.
- Matlab
>> demo
The result will be printed like
Accuracy (std) of GDT vs. DT:
GDT 0.75 (0.10)
DT 0.50 (0.11)
It may vary for each run because the random selection of training samples.
Z. Zhang, Y. Song, H. Cui, J. Wu, F. Schwartz, and H. Qi. "Topological Analysis and Gaussian Decision Tree: Effective Representation and Classification of Biosignals of Small Sample Size". IEEE Transactions on Biomedical Engineering (TBME), 2016. [PDF]
Z. Zhang, Y. Song, H. Cui, J. Wu, F. Schwartz, and H. Qi. "Early Mastitis Diagnosis through Topological Analysis of Biosignalsfrom Low-Voltage Alternate Current Electrokinetics". International Conference on the IEEE Engineering in Medicine and Biology Society (EMBC), 2015. [PDF]
@inproceedings{zhang2015early,
title={Early mastitis diagnosis through topological analysis of biosignals from low-voltage alternate current electrokinetics},
author={Zhang, Zhifei and Song, Yang and Cui, Haochen and Wu, Jayne and Schwartz, Fernando and Qi, Hairong},
booktitle={2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)},
pages={542--545},
year={2015},
organization={IEEE}
}