Research Article
An Ensemble Method for High-Dimensional Multilabel Data
Algorithm 1
Ensemble multilabel classifier using feature selection (EMCFS).
Input: : The training multilabel data set; | : The parameter of ; | : The test data set without labels; | Output: : The label set of ; | Training stage | For each label in do | Obtain and of from according to Def. (1) and (2); | Calculate with and according to Def. (6); | For each feature in do | Calculate and according to Def. (4) and (5); | Select the features whose or is larger than ; | Train a binary classifier on with the selected features; | Endfor | Prediction stage | For each sample in do | ; | For each classifier do | , where returns 0 or 1; | Endfor |
|