Filters








36,923 Hits in 6.3 sec

Feature Ranking for Hierarchical Multi-Label Classification with Tree Ensemble Methods

Matej Petković, Sašo Džeroski, Dragi Kocev
2020 Acta Polytechnica Hungarica  
In this work, we address the task of feature ranking for hierarchical multi-label classification (HMLC).  ...  Here, we propose a group of feature ranking methods based on three established ensemble methods of predictive clustering trees: Bagging, Random Forests and Extra Trees.  ...  When coupled with the suitable ensemble method, all three scores outperform the HMLC-Relief feature ranking.  ... 
doi:10.12700/aph.17.10.2020.10.8 fatcat:5kdgkh3qpjbajif7ffo2scajwe

A Triple-Random Ensemble Classification Method for Mining Multi-label Data

Gulisong Nasierding, Abbas Z. Kouzani, Grigorios Tsoumakas
2010 2010 IEEE International Conference on Data Mining Workshops  
This paper presents a triple-random ensemble learning method for handling multi-label classification problems.  ...  The proposed method integrates and develops the concepts of random subspace, bagging and random k-labelsets ensemble learning methods to form an approach to classify multi-label data.  ...  CONCLUSION The paper presented a new ensemble learning method, named triple-random ensemble classification for dealing with multi-label learning problems.  ... 
doi:10.1109/icdmw.2010.139 dblp:conf/icdm/NasierdingKT10 fatcat:abtnzghjjne7bhnhhzwm4uboc4

FCBF3Rules: A feature selection method for multi-label datasets

Shima Kashef, Hossein Nezamabadi-pour, Bahareh Nikpour
2018 2018 3rd Conference on Swarm Intelligence and Evolutionary Computation (CSIEC)  
Multi-label classification has gained significant attention during recent years, due to the increasing number of modern applications associated with multi-label data.  ...  LIFT is a multi-label classifier which utilizes a new strategy to multi-label learning by leveraging label-specific features.  ...  Ensemble When BR transforms the multi-label dataset into q binary single-label datasets, the ensemble on n filter feature selection methods is performed on these datasets.  ... 
doi:10.1109/csiec.2018.8405419 fatcat:3ndtzqwijzeqjgrymh5eer6vby

MLIFT: Enhancing Multi-label Classifier with Ensemble Feature Selection

Sh kashef, H. Nezamabadi-pour
2019 Journal of Artificial Intelligence and Data Mining  
Multi-label classification has gained significant attention during recent years, due to the increasing number of modern applications associated with multi-label data.  ...  LIFT is a multi-label classifier which utilizes a new strategy to multi-label learning by leveraging label-specific features.  ...  Algorithm 1: the proposed method Input: Multi-label dataset D with M features, N samples and q labels Output: selected features F 1: Transform D to q single-label datasets using BR 2: for i = 1:q 3: Filter1  ... 
doi:10.22044/jadm.2018.5780.1688 doaj:8d31b19f77964bb4993cf84c3864cfae fatcat:fasr3kivffhmdiw7hxtpol3jzi

Multi-Label Bioinformatics Data Classification With Ensemble Embedded Feature Selection

Yumeng Guo, Fulai Chung, Guozheng Li, Lei Zhang
2019 IEEE Access  
However, most of the proposed methods, which aimed at dealing with multilabel feature selection problem in the past few years, only adopt a simple and direct strategy that transforms the multi-label feature  ...  Furthermore, it can reduce the accumulated errors of data itself by employing an ensemble method.  ...  Ensemble Embedded Feature Selection, which can deal with the multi-label feature selection problems in bioinformatics data.  ... 
doi:10.1109/access.2019.2931035 fatcat:7roc2dqx7rdn7dhc7hctmerk6a

A Novel Multi-Feature Joint Learning Ensemble Framework for Multi-Label Facial Expression Recognition

Wanzhao Li, Mingyuan Luo, Peng Zhang, Wei Huang
2021 IEEE Access  
The proposed framework combines global features with several different local key features to consider the multiple labels of expressions embodied in many facial action units.  ...  To overcome this challenge, a novel multi-feature joint learning ensemble framework, called MF-JLE framework, is proposed.  ...  Finally, multi-feature joint learning ensemble (MF-JLE) framework achieves better performance and outperforms some state-of-the-art multi-label facial expression recognition methods in most criteria by  ... 
doi:10.1109/access.2021.3108838 fatcat:3zte75g56rf3zhhps4ffdlytp4

Deep tree-ensembles for multi-output prediction [article]

Felipe Kenji Nakano, Konstantinos Pliakos, Celine Vens
2021 arXiv   pre-print
In this direction, we propose a novel deep tree-ensemble (DTE) model, where every layer enriches the original feature set with a representation learning component based on tree-embeddings.  ...  Despite that, these approaches simply employ label classification probabilities as induced features and primarily focus on traditional classification and regression tasks, leaving multi-output prediction  ...  More precisely, it enforces local low-rank on predictions with the same labels, whereas the rank is expanded for predictions with different labels.  ... 
arXiv:2011.02829v2 fatcat:44jd5rm6kbcdhpkbqbm5isfl2q

Feature Ranking for Semi-supervised Learning [article]

Matej Petković, Sašo Džeroski, Dragi Kocev
2020 arXiv   pre-print
The feature rankings are learned in the context of classification and regression as well as in the context of structured output prediction (multi-label classification, hierarchical multi-label classification  ...  To address these challenges, we propose semi-supervised learning of feature ranking.  ...  Different SSL feature ranking methods perform the best for the different tasks: Symbolic ranking is the best for the regression and multi-target regression, Random forest ranking for multi-label classification  ... 
arXiv:2008.03937v1 fatcat:7nipvmrnf5fyto6h24kpyprhh4

Empirical Study of Multi-label Classification Methods for Image Annotation and Retrieval

Gulisong Nasierding, Abbas Z. Kouzani
2010 2010 International Conference on Digital Image Computing: Techniques and Applications  
This paper presents an empirical study of multi-label classification methods, and gives suggestions for multi-label classification that are effective for automatic image annotation applications.  ...  The study shows that triple random ensemble multi-label classification algorithm (TREMLC) outperforms among its counterparts, especially on scene image dataset.  ...  Although RAkEL, EPS and TREMLC are effective multi-label ensemble learning methods, however, they pursuing different subsets selection schemes for building multi-label ensemble classifiers. III.  ... 
doi:10.1109/dicta.2010.113 dblp:conf/dicta/NasierdingK10 fatcat:6bpjdbth35adxknxauedb2thoe

Predicting drug side effects by multi-label learning and ensemble learning

Wen Zhang, Feng Liu, Longqiang Luo, Jingxia Zhang
2015 BMC Bioinformatics  
Methods: In this paper, we propose a novel method 'feature selection-based multi-label k-nearest neighbor method' (FS-MLKNN), which can simultaneously determine critical feature dimensions and construct  ...  When compared with other state-of-the-art methods, the ensemble method produces better performances on benchmark datasets.  ...  Feature selection-based multi-label k-nearest neighbor method We design the feature selection-based multi-label k-nearest neighbor method (FS-MLKNN) to simultaneously determine the optimal feature dimensions  ... 
doi:10.1186/s12859-015-0774-y pmid:26537615 pmcid:PMC4634905 fatcat:zpmct7qr3nhbjgt65uuu5wtpqe

Guest editors' introduction to the special issue on Discovery Science

Larisa Soldatova, Joaquin Vanschoren
2020 Machine Learning  
Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.  ...  Petkovic et al. present their work on Multi-label feature ranking with ensemble methods.  ...  They propose three ensemble-based feature ranking scores for multi-label classification, and demonstrate empirically that the proposed ranking scores outperform current stateof-the-art methods in the quality  ... 
doi:10.1007/s10994-020-05922-3 fatcat:57jmykowdbdunfthtwcg2hizkm

A Dynamic Two-Layers MI and Clustering-based Ensemble Feature Selection for Multi-Labels Text Classification

Adil Yaseen Taha, Sabrina Tiun, Abdul Hadi, Masri Ayob, Ali Sabah
2020 International Journal of Advanced Computer Science and Applications  
In addition, this paper proposes an enhanced FS method called dynamic multi-label two-layers MI and clusteringbased ensemble feature selection algorithm (DMMC-EFS).  ...  With regards to handling correlation and high dimensionality problems in multi-label text classification, this paper investigates the various heterogeneous FS ensemble schemes.  ...  the multi-label ensemble FS methods: multi-label mean-based ensemble feature (ME-mean) selection method and multi-label plurality vote ensemble FS method (ME-PV) with our proposed method, dynamic multi-label  ... 
doi:10.14569/ijacsa.2020.0110764 fatcat:uj4mbpalqfhzljp6b6fcq4pcgu

Multi-label ensemble based on variable pairwise constraint projection

Ping Li, Hong Li, Min Wu
2013 Information Sciences  
Empirical studies have shown the superiority of the proposed method in comparison with other approaches.  ...  To achieve these goals, this paper presents a novel multi-label classification framework named Variable Pairwise Constraint projection for Multi-label Ensemble (VPCME).  ...  Besides the above classical methods, other ensemble methods such as decision tree ensemble [1] , active ensemble [22] , dynamic ensemble selection [43] , feature sets ensemble [44] and artificial  ... 
doi:10.1016/j.ins.2012.07.066 fatcat:uob6q5joizcp7azxdynmcw5xou

An extensive experimental comparison of methods for multi-label learning

Gjorgji Madjarov, Dragi Kocev, Dejan Gjorgjevikj, Sašo Džeroski
2012 Pattern Recognition  
Furthermore, RF-PCT exhibited the best performance according to all measures for multi-label ranking.  ...  Multi-label learning has received significant attention in the research community over the past few years: this has resulted in the development of a variety of multi-label learning methods.  ...  In this study, we extend this categorization of multi-label methods with a third group of methods, namely, ensemble methods.  ... 
doi:10.1016/j.patcog.2012.03.004 fatcat:wcjofxautffxlajqwh64xwwd3y

Improving Multilabel Classification Performance by Using Ensemble of Multi-label Classifiers [chapter]

Muhammad Atif Tahir, Josef Kittler, Krystian Mikolajczyk, Fei Yan
2010 Lecture Notes in Computer Science  
multi-label methods by ensemble techniques.  ...  Recently, a considerable amount of research has been concerned with the development of "good" multi-label learning methods.  ...  This method was shown to perform well against other multi-label classifiers but mainly on ranking measures.  ... 
doi:10.1007/978-3-642-12127-2_2 fatcat:wf25oxjesfbr7ahqbhit6pxejy
« Previous Showing results 1 — 15 out of 36,923 results