将这两者结合起来,就形成了多标签信息驱动特征选择(Multi-label Informed Feature Selection)。 多标签问题简介: 传统的分类问题通常是单标签分类,即每个样本只能属于一个类别。而在现实场景中,很多问题都是多标签问题,一个样本可能属于多个不同的类别。例如,在图像分类中,一张图像可能同时包含猫和树,而不是仅仅属于...
Label relationshipsMulti-label feature selection is an efficient technique to alleviate the high dimensionality for multi-label learning. Existing multi-label feature selection methods based on information theory either deal with labels individually or treat all label relationships as redundancy. However, ...
multi-label data are obtained based on granular computing; second, the feature complementarity is estimated based on neighborhood mutual information without ... W Qian,X Long,Y Wang,... - 《Applied Soft Computing》 被引量: 0发表: 2020年 Multilabel feature selection using ML-ReliefF and neighb...
Information theoretical based methods have attracted a great attention in recent years, and gained promising results to deal with multi-label data with high dimensionality. However, most of the existing methods are either directly transformed from heuristic single-label feature selection methods or ineffi...
Feature selection, as an important pre-processing technique, can efficiently mitigate the issue of “the curse of dimensionality” by selecting discriminative features especially for multi-label learning, a discriminative feature subset can improve the classification accuracy. The existing feature selection ...
Proposed is a new multi-label feature selection method that captures relationships between features and labels without transforming the problem into single-label classification. Using approximated joint mutual information, the proposed incremental feature selection algorithm provides markedly better classification...
labelfeaturemultiselectioninformedmifs Multi-LabelInformedFeatureSelectionLingJian1,2∗,JundongLi1∗,KaiShu1,HuanLiu11.ComputerScienceandEngineering,ArizonaStateUniversity,Tempe,85281,USA2.CollegeofScience,ChinaUniversityofPetroleum,Qingdao,266555,China{ling.jian,jundong.li,kai.shu,huan.liu}@asu.eduAbstra...
Semi-supervised learning and multi-label learning pose different challenges for feature selection, which is one of the core techniques for dimension reduction, and the exploration of reducing feature space for multi-label learning with incomplete label information is far from satisfactory. Existing featur...
Multi-label learning generalizes traditional learning by allowing an instance to belong to multiple labels simultaneously. This causes multi-label data to be characterized by its large label space dimensionality and the dependencies among labels. These challenges have been addressed by feature selection te...
multi-label naive Bayes, is proposed. In order to improve its performance, a two-stage filter-wrapper feature selection strategy is also incorporated. Specifically, in the first stage, feature extraction techniques based on principle component analysis (PCA) are used to eliminate irrelevant and ...