Title :
Use of unlabeled samples for mitigating the Hughes phenomenon
Author :
Shahshahani, Behzad M. ; Landgrebe, David A.
Author_Institution :
Sch. of Electr. Eng., Purdue Univ., West Lafayette, IN, USA
Abstract :
The use of unlabeled samples in improving the performance of classifiers is studied. When the number of training samples is fixed and small, additional feature measurements may reduce the performance of a statistical classifier. It is shown that by using unlabeled samples estimates of the parameters can be improved and therefore this phenomenon may be mitigated. Various methods for using unlabeled samples are reviewed and experimental results are provided
Keywords :
geophysical techniques; geophysics computing; image recognition; remote sensing; Hughes phenomenon; classifier; geophysical measurement technique; image classification; land surface remote sensing; performance; statistical classifier; training sample; unlabeled sample; Covariance matrix; Degradation; Layout; NASA; Parameter estimation; Probability density function; Production; Sensor phenomena and characterization; Taylor series;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 1993. IGARSS '93. Better Understanding of Earth Environment., International
Conference_Location :
Tokyo
Print_ISBN :
0-7803-1240-6
DOI :
10.1109/IGARSS.1993.322762
