Multi-Label Learning With Hidden Labels
In multi-label learning, each object is represented by a single instance and associated with multiple labels simultaneously. Existing multi-label learning approaches mainly construct classification models with a fixed set of target labels (observed labels). However, in the big data era, it is diffic...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8988239/ |
| _version_ | 1818557920624771072 |
|---|---|
| author | Jun Huang Haowei Rui Guorong Li Xiwen Qu Tao Tao Xiao Zheng |
| author_facet | Jun Huang Haowei Rui Guorong Li Xiwen Qu Tao Tao Xiao Zheng |
| author_sort | Jun Huang |
| collection | DOAJ |
| description | In multi-label learning, each object is represented by a single instance and associated with multiple labels simultaneously. Existing multi-label learning approaches mainly construct classification models with a fixed set of target labels (observed labels). However, in the big data era, it is difficult to provide a fully complete label set for a data set. In some real applications, there are multiple labels hidden in the data set, especially for those large-scale data sets. In this paper, a novel approach named MLLHL is proposed to not only discover the hidden labels in the training data but also predict these hidden labels and observed labels for unseen examples simultaneously. We assume that the observed labels are just a subset of labels which are selected from the full label set, and the rest ones are omitted by the annotators during the labeling stage. Extensive experiments show the competitive performance of MLLHL against other state-of-the-art multi-label learning approaches. |
| first_indexed | 2024-12-14T00:06:01Z |
| format | Article |
| id | doaj.art-b2a6a8aac9404bc8aa05dfd80b61224a |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T00:06:01Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-b2a6a8aac9404bc8aa05dfd80b61224a2022-12-21T23:26:02ZengIEEEIEEE Access2169-35362020-01-018296672967610.1109/ACCESS.2020.29725998988239Multi-Label Learning With Hidden LabelsJun Huang0https://orcid.org/0000-0002-2022-5747Haowei Rui1https://orcid.org/0000-0001-5349-9392Guorong Li2https://orcid.org/0000-0003-3954-2387Xiwen Qu3https://orcid.org/0000-0002-3912-4831Tao Tao4https://orcid.org/0000-0002-5630-2070Xiao Zheng5https://orcid.org/0000-0001-7565-7072School of Computer Science and Technology, Anhui University of Technology, Ma’anshan, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan, ChinaSchool of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan, ChinaSchool of Computer Science and Technology, Anhui University of Technology, Ma’anshan, ChinaIn multi-label learning, each object is represented by a single instance and associated with multiple labels simultaneously. Existing multi-label learning approaches mainly construct classification models with a fixed set of target labels (observed labels). However, in the big data era, it is difficult to provide a fully complete label set for a data set. In some real applications, there are multiple labels hidden in the data set, especially for those large-scale data sets. In this paper, a novel approach named MLLHL is proposed to not only discover the hidden labels in the training data but also predict these hidden labels and observed labels for unseen examples simultaneously. We assume that the observed labels are just a subset of labels which are selected from the full label set, and the rest ones are omitted by the annotators during the labeling stage. Extensive experiments show the competitive performance of MLLHL against other state-of-the-art multi-label learning approaches.https://ieeexplore.ieee.org/document/8988239/Multi-label learningdiscovering hidden labels |
| spellingShingle | Jun Huang Haowei Rui Guorong Li Xiwen Qu Tao Tao Xiao Zheng Multi-Label Learning With Hidden Labels IEEE Access Multi-label learning discovering hidden labels |
| title | Multi-Label Learning With Hidden Labels |
| title_full | Multi-Label Learning With Hidden Labels |
| title_fullStr | Multi-Label Learning With Hidden Labels |
| title_full_unstemmed | Multi-Label Learning With Hidden Labels |
| title_short | Multi-Label Learning With Hidden Labels |
| title_sort | multi label learning with hidden labels |
| topic | Multi-label learning discovering hidden labels |
| url | https://ieeexplore.ieee.org/document/8988239/ |
| work_keys_str_mv | AT junhuang multilabellearningwithhiddenlabels AT haoweirui multilabellearningwithhiddenlabels AT guorongli multilabellearningwithhiddenlabels AT xiwenqu multilabellearningwithhiddenlabels AT taotao multilabellearningwithhiddenlabels AT xiaozheng multilabellearningwithhiddenlabels |