文章基本信息

标题：THE QUALITY DETERMINATION OF COCONUT WOOD DENSITY USING LEARNING VECTOR QUANTIZATION
本地全文：下载
作者：IGN. NGESTI YUWONO ; RICARDUS ANGGI PRAMUNENDAR ; PULUNG NURTANTIO ANDONO 等
期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2013
卷号：57
期号：1
出版社：Journal of Theoretical and Applied
摘要：Coconut wood is one of the wood species used in Indonesia as a material for furniture and building construction. Coconut wood quality in terms of strength and durability is determined by many factors, such as a vascullar bundle density pattern on the wood. Currently the quality determination of coconut wood were done visually by the expert by looking at the pattern density vascullar bundle on coconut wood. The coconut wood qualities visually were classified into patterns of low density, low quality coconut wood; patterns of medium density, medium quality coconut wood; patterns of high density, high quality coconut wood. The classification of coconut wood density pattern could be done by applying intelligent technique, such as Learning Vector Quatization (LVQ). The LVQ model is a classification method thah uses a competitive supervised learning algorithm. The experimental result showed that the accuracy was 65.7% with kappa value of 0.481. the obtained result had a sufficient congruence between assessment and appraisal target density using LVQ.
关键词：Coconut Wood Density; LVQ