《分类高级课题AdvancedTopicsonClassification教学课堂PPT》由会员分享，可在线阅读，更多相关《分类高级课题AdvancedTopicsonClassification教学课堂PPT（22页珍藏版）》请在金锄头文库上搜索。
1、Advanced Topics on ClassificationQuan Zou (邹 权) (Ph.D.& Assistant Professor)8/28/20241OutlineImbalance Binary ClassificationMulti Class, Multi Label ClassificationMulti Instance ClassificationSemi-supervised and Transductive ClassificationEnsemble LearningOthers8/28/20242Imbalance binary classificat
2、ionApplication：Credit Card CheatSpam IdentificationFinding OilBioinformatics8/28/20243Imbalance binary classificationStrategy of samplingOver-samplingUnder-samplingRandom-samplingSpecial-sampling (SMOTE)Strategy of costEqual to aboveOne-class leaning8/28/20244Multi Class, Multi LabelMulti ClassOne v
3、s One (time consuming)One vs All (imbalance)Tree Multi LabelJRS (http:/tunedit.org/challenge/JRS12Contest)Text, Image ClassificationKNNmeka, mulan8/28/20245mulan8/28/202468/28/20247meka8/28/20248Multi Instance ClassificationDrug Design, Image UnderstandingPackage, Instance DD8/28/202498/28/202410Sem
4、i-supervised and Transductive ClassificationSemi-supervised ClassificationUnlabeled samples are importantCo-training and Tri-training8/28/202411Transductive Classification8/28/2024128/28/202413Ensemble learningBagging8/28/202414Ensemble learningBoosting8/28/202415Ensemble learningRandom Forest8/28/2
5、02416Ensemble learning for Class Imbalance Problem8/28/2024178/28/202418StrategyFirst, the negative set is divided randomly into several subsets equally. Every subset together with the positive set is a class balance training set. Then several different classifiers are selected and trained with thes
6、e balance training sets. They will vote for the last prediction when facing new samples.The samples will be added to the next two classifiers training sets if they are misclassified.Reference邹权, 郭茂祖, 刘扬, 王峻. 类别不平衡的分类方法及在生物信息学中的应用. 计算机研究与发展. 2010,47(8):1407-1414 X.-Y. Liu, J. Wu, and Z.-H. Zhou. Expl
7、oratory undersampling for class-imbalance learning. IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, 2009, 39(2): 539-550 8/28/202419OthersActive learningLazy learningParallel learning (mahout)OptimizationFeatures Selection (GA)Parameters Tune (Grid, PSO)8/28/202420Email: 8/28/202421 素材和资料部分来自素材和资料部分来自网络，如有帮助请下载网络，如有帮助请下载!
