The research of semi-supervised manifold learning algorithm based on markov property
QingSong XU, Liang Shen, Dongsheng Cao, Xin HUANG · Scientia Sinica Mathematica · 2015
On the background of complex data in pharmacokinetics, we use manifold classification algorithm which has Markov property to model the QSAR. Simulation study show that when there exists some manifold structures, the manifold classification algorithm can give better results of classification. Three real datasets related to different categorical bioactivities of compounds are used to test the performance of MSSC. Statistical evaluation shows that the proposed algorithm is a promising and alternative modeling approach compared to the baseline models.