西南师范大学学报(自然科学版)
西南師範大學學報(自然科學版)
서남사범대학학보(자연과학판)
JOURNAL OF SOUTHWEST CHINA NORMAL UNIVERSITY
2014年
4期
124-128
,共5页
k部排序算法%半监督学习%正则化瑞利系数%本体
k部排序算法%半鑑督學習%正則化瑞利繫數%本體
k부배서산법%반감독학습%정칙화서리계수%본체
k-partite ranking algorithm%semi-supervised learning%normalized Rayleigh coefficient%ontology
借鉴 Ralaivola给出的基于正则化瑞利系数的半监督二部排序学习算法的思想,提出基于正则化瑞利系数的半监督k 部排序学习算法。同时,将此方法运用于本体相似度计算和本体映射。通过两个实验说明新算法是有效的。
藉鑒 Ralaivola給齣的基于正則化瑞利繫數的半鑑督二部排序學習算法的思想,提齣基于正則化瑞利繫數的半鑑督k 部排序學習算法。同時,將此方法運用于本體相似度計算和本體映射。通過兩箇實驗說明新算法是有效的。
차감 Ralaivola급출적기우정칙화서리계수적반감독이부배서학습산법적사상,제출기우정칙화서리계수적반감독k 부배서학습산법。동시,장차방법운용우본체상사도계산화본체영사。통과량개실험설명신산법시유효적。
By means of semi-supervised bipartite ranking based on the normalized Rayleigh coefficient, which was raised by Ralaivola,a new algorithm for semi-supervised k-partite ranking has been propose on the basis of the normalized Rayleigh coefficient.At the same time,this method is applied to the ontology similarity measure and ontologies mapping.The two experiments show that the new algorithm is efficient.