Application of the relevance vector machine to drought monitoring
Soft Computing for Problem Solving: SocProS 2017, Volume 1, 2019•Springer
The study demonstrates the application of relevance vector machines (RVMs) to drought
monitoring, specifically, agricultural drought classification. The model is based on a crop
water stress function that serves as an indicator of agricultural drought in the study area. The
RVM framework performs a multi-class classification on the crop stress feature vector and
yields probabilistic classification of drought classes. The results indicate that the uncertainty
involved in classification is known with the help of the RVM-based classification model.
monitoring, specifically, agricultural drought classification. The model is based on a crop
water stress function that serves as an indicator of agricultural drought in the study area. The
RVM framework performs a multi-class classification on the crop stress feature vector and
yields probabilistic classification of drought classes. The results indicate that the uncertainty
involved in classification is known with the help of the RVM-based classification model.
Abstract
The study demonstrates the application of relevance vector machines (RVMs) to drought monitoring, specifically, agricultural drought classification. The model is based on a crop water stress function that serves as an indicator of agricultural drought in the study area. The RVM framework performs a multi-class classification on the crop stress feature vector and yields probabilistic classification of drought classes. The results indicate that the uncertainty involved in classification is known with the help of the RVM-based classification model.
Springer
Showing the best result for this search. See all results