Coherent hierarchical multi-label classification networks
Hierarchical multi-label classification (HMC) is a challenging classification task extending standard multi-label classification problems by imposing a hierarchy constraint on the classes. In this paper, we propose C-HMCNN(h), a novel approach for HMC problems, which, given a network h for the under...
| Main Authors: | , |
|---|---|
| Format: | Conference item |
| Language: | English |
| Published: |
NeurIPS
2020
|
| _version_ | 1826265020385722368 |
|---|---|
| author | Giunchiglia, E Lukasiewicz, T |
| author_facet | Giunchiglia, E Lukasiewicz, T |
| author_sort | Giunchiglia, E |
| collection | OXFORD |
| description | Hierarchical multi-label classification (HMC) is a challenging classification task extending standard multi-label classification problems by imposing a hierarchy constraint on the classes. In this paper, we propose C-HMCNN(h), a novel approach for HMC problems, which, given a network h for the underlying multi-label classification problem, exploits the hierarchy information in order to produce predictions coherent with the constraint and improve performance. We conduct an extensive experimental analysis showing the superior performance of C-HMCNN(h) when compared to state-of-the-art models. |
| first_indexed | 2024-03-06T20:17:06Z |
| format | Conference item |
| id | oxford-uuid:2c7d5a86-7f2f-434a-b58b-3e8686b6165a |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T20:17:06Z |
| publishDate | 2020 |
| publisher | NeurIPS |
| record_format | dspace |
| spelling | oxford-uuid:2c7d5a86-7f2f-434a-b58b-3e8686b6165a2022-03-26T12:37:33ZCoherent hierarchical multi-label classification networksConference itemhttp://purl.org/coar/resource_type/c_5794uuid:2c7d5a86-7f2f-434a-b58b-3e8686b6165aEnglishSymplectic ElementsNeurIPS2020Giunchiglia, ELukasiewicz, THierarchical multi-label classification (HMC) is a challenging classification task extending standard multi-label classification problems by imposing a hierarchy constraint on the classes. In this paper, we propose C-HMCNN(h), a novel approach for HMC problems, which, given a network h for the underlying multi-label classification problem, exploits the hierarchy information in order to produce predictions coherent with the constraint and improve performance. We conduct an extensive experimental analysis showing the superior performance of C-HMCNN(h) when compared to state-of-the-art models. |
| spellingShingle | Giunchiglia, E Lukasiewicz, T Coherent hierarchical multi-label classification networks |
| title | Coherent hierarchical multi-label classification networks |
| title_full | Coherent hierarchical multi-label classification networks |
| title_fullStr | Coherent hierarchical multi-label classification networks |
| title_full_unstemmed | Coherent hierarchical multi-label classification networks |
| title_short | Coherent hierarchical multi-label classification networks |
| title_sort | coherent hierarchical multi label classification networks |
| work_keys_str_mv | AT giunchigliae coherenthierarchicalmultilabelclassificationnetworks AT lukasiewiczt coherenthierarchicalmultilabelclassificationnetworks |