Title :
The Improvement of Transductive Support Vector Machine and Its Application to the Environmental Monitoring of Greenhouse Plant Growth
Author_Institution :
Dept. of Math., Tangshan Teacher´´s Coll., Tangshan, China
Abstract :
The study on Transductive Support Vector Machine (TSVM) has made little progress since Vapnik has put forth the concept in the late 1990s, as the algorithm for TSVM optimization model cannot be easily found. Here we try to transform the problem of TSVM optimization into an unconstrained one and then to construct the smooth unconstrained optimization with a kernel. Based on this optimization, a TSVM of which optimization problem is easier to solve is devised to break through the bottleneck of deepening the research on TSVM. Furthermore, this TSVM is applied to the environmental monitoring of greenhouse plant growth as a new method.
Keywords :
computerised monitoring; greenhouses; optimisation; support vector machines; TSVM optimization model; environmental monitoring; greenhouse plant growth; transductive support vector machine; Air pollution; Green products; Kernel; Monitoring; Optimization; Support vector machines; Training; Environmental monitoring; Greenhouse plant; Growing environment; Optimization; Transductive Support Vector Machine; Unconstrained Optimization;
Conference_Titel :
Cryptography and Network Security, Data Mining and Knowledge Discovery, E-Commerce & Its Applications and Embedded Systems (CDEE), 2010 First ACIS International Symposium on
Conference_Location :
Qinhuangdao
Print_ISBN :
978-1-4244-9595-5
DOI :
10.1109/CDEE.2010.83
