random-forest

The project is mainly to study the state-of-the-art algorithm, called random forest, and applicates it to chemometrics, chemoinformatics, and bioinformatics. http://www.stat.berkeley.edu/~breiman/RandomForests/ (1) Variable importance sampling-based adaptive random forest as a useful tool to screen underlying lead compounds Good performance of ensemble approaches could generally be obtained when base classifiers are diverse and accurate. In the present study, variable importance sampling-based adaptive random forest (visaRF) was proposed to obtain superior classification performance to the primal one-step RF. visaRF takes a convenient, yet very effective way, called variable importance sampling (VIS), to select the more eligible variable subset at each splitting node instead of simple random sampling and thereby strengthen the accuracy of individual trees, without sacrificing diversity between them. Additionally, the iterative use of variable