Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation
In order to enrich the expression of single-view features and realize complementary learning between multiple views, a multi-view latent space fusion representation method based on enhancing feature deep learning was proposed. The model has three submodules: single-view enhanced learning, multi-view...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Editorial Office of Journal of Taiyuan University of Technology
2022-07-01
|
| Series: | Taiyuan Ligong Daxue xuebao |
| Subjects: | |
| Online Access: | https://tyutjournal.tyut.edu.cn/englishpaper/show-1903.html |
| _version_ | 1797208285091201024 |
|---|---|
| author | Tingyi ZHENG Jiaqi WU Binbin ZHANG Li WANG |
| author_facet | Tingyi ZHENG Jiaqi WU Binbin ZHANG Li WANG |
| author_sort | Tingyi ZHENG |
| collection | DOAJ |
| description | In order to enrich the expression of single-view features and realize complementary learning between multiple views, a multi-view latent space fusion representation method based on enhancing feature deep learning was proposed. The model has three submodules: single-view enhanced learning, multi-view complementary fusion, and self-representation based on clustering task. First of all, the dynamic routing mechanism of capsule network is introduced, and the interval loss penalty item is added to the objective function to obtain the single-view feature of differential features enhancement. Next, the important features of different views are fused, the public latent space of multiple views is learned, the complementary representation between view features is realized, and the classification task is met. Then, the subspace clustering algorithm is used to learn the self-representation matrix of latent space, and the low-rank representation constraints of the latent space reconstruction error matrix and the noise data matrix are added to the objective function to obtain the fusion representation that meets the clustering task. Finally, the classification and clustering experiments were conducted on four data sets. Compared with multiple benchmark algorithms, this algorithm has been steadily improved in performace, and the learned fusion characterization can better meet the needs of downstream classification and clustering tasks. |
| first_indexed | 2024-04-24T09:36:22Z |
| format | Article |
| id | doaj.art-5fb6b2b2fb374f9db379020637c5bac6 |
| institution | Directory Open Access Journal |
| issn | 1007-9432 |
| language | English |
| last_indexed | 2024-04-24T09:36:22Z |
| publishDate | 2022-07-01 |
| publisher | Editorial Office of Journal of Taiyuan University of Technology |
| record_format | Article |
| series | Taiyuan Ligong Daxue xuebao |
| spelling | doaj.art-5fb6b2b2fb374f9db379020637c5bac62024-04-15T09:15:45ZengEditorial Office of Journal of Taiyuan University of TechnologyTaiyuan Ligong Daxue xuebao1007-94322022-07-0153469770610.16355/j.cnki.issn1007-9432tyut.2022.04.0141007-9432(2022)04-0697-10Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion RepresentationTingyi ZHENG0Jiaqi WU1Binbin ZHANG2Li WANG3College of Information and Computer Science, Taiyuan University of Technology, Taiyuan 030024, ChinaUniversity of New South Wales, Sydney 1466, AustraliaCollege of Data Science, Taiyuan University of Technology, Taiyuan 030024, ChinaCollege of Data Science, Taiyuan University of Technology, Taiyuan 030024, ChinaIn order to enrich the expression of single-view features and realize complementary learning between multiple views, a multi-view latent space fusion representation method based on enhancing feature deep learning was proposed. The model has three submodules: single-view enhanced learning, multi-view complementary fusion, and self-representation based on clustering task. First of all, the dynamic routing mechanism of capsule network is introduced, and the interval loss penalty item is added to the objective function to obtain the single-view feature of differential features enhancement. Next, the important features of different views are fused, the public latent space of multiple views is learned, the complementary representation between view features is realized, and the classification task is met. Then, the subspace clustering algorithm is used to learn the self-representation matrix of latent space, and the low-rank representation constraints of the latent space reconstruction error matrix and the noise data matrix are added to the objective function to obtain the fusion representation that meets the clustering task. Finally, the classification and clustering experiments were conducted on four data sets. Compared with multiple benchmark algorithms, this algorithm has been steadily improved in performace, and the learned fusion characterization can better meet the needs of downstream classification and clustering tasks.https://tyutjournal.tyut.edu.cn/englishpaper/show-1903.htmlmulti-view learninglatent space learningsubspace clustering |
| spellingShingle | Tingyi ZHENG Jiaqi WU Binbin ZHANG Li WANG Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation Taiyuan Ligong Daxue xuebao multi-view learning latent space learning subspace clustering |
| title | Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation |
| title_full | Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation |
| title_fullStr | Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation |
| title_full_unstemmed | Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation |
| title_short | Enhancing Feature Deep Learning to Improve Multi-view Latent Space Fusion Representation |
| title_sort | enhancing feature deep learning to improve multi view latent space fusion representation |
| topic | multi-view learning latent space learning subspace clustering |
| url | https://tyutjournal.tyut.edu.cn/englishpaper/show-1903.html |
| work_keys_str_mv | AT tingyizheng enhancingfeaturedeeplearningtoimprovemultiviewlatentspacefusionrepresentation AT jiaqiwu enhancingfeaturedeeplearningtoimprovemultiviewlatentspacefusionrepresentation AT binbinzhang enhancingfeaturedeeplearningtoimprovemultiviewlatentspacefusionrepresentation AT liwang enhancingfeaturedeeplearningtoimprovemultiviewlatentspacefusionrepresentation |