A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets
Multilabel classification is a recently conceptualized task in machine learning. Contrary to most of the research that has so far focused on classification machinery, we take a data-centric approach and provide an integrative framework that blends qualitative and quantitative descriptions of multila...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/12/2/346 |
| _version_ | 1797343021210009600 |
|---|---|
| author | Francisco J. Valverde-Albacete Carmen Peláez-Moreno |
| author_facet | Francisco J. Valverde-Albacete Carmen Peláez-Moreno |
| author_sort | Francisco J. Valverde-Albacete |
| collection | DOAJ |
| description | Multilabel classification is a recently conceptualized task in machine learning. Contrary to most of the research that has so far focused on classification machinery, we take a data-centric approach and provide an integrative framework that blends qualitative and quantitative descriptions of multilabel data sources. By combining lattice theory, in the form of formal concept analysis, and entropy triangles, obtained from information theory, we explain from first principles the fundamental issues of multilabel datasets such as the dependencies of the labels, their imbalances, or the effects of the presence of hapaxes. This allows us to provide guidelines for resampling and new data collection and their relationship with broad modelling approaches. We have empirically validated our framework using 56 open datasets, challenging previous characterizations that prove that our formalization brings useful insights into the task of multilabel classification. Further work will consider the extension of this formalization to understand the relationship between the data sources, the classification methods, and ways to assess their performance. |
| first_indexed | 2024-03-08T10:41:38Z |
| format | Article |
| id | doaj.art-c7875c159a3a49419a5c5295c9f5b792 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-08T10:41:38Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-c7875c159a3a49419a5c5295c9f5b7922024-01-26T17:34:19ZengMDPI AGMathematics2227-73902024-01-0112234610.3390/math12020346A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning DatasetsFrancisco J. Valverde-Albacete0Carmen Peláez-Moreno1Department of Signal Theory and Communications, Telematic Systems and Computation, Universidad Rey Juan Carlos, 28942 Fuenlabrada, Madrid, SpainDepartment of Signal Theory and Communications, Universidad Carlos III de Madrid, 28911 Leganés, Madrid, SpainMultilabel classification is a recently conceptualized task in machine learning. Contrary to most of the research that has so far focused on classification machinery, we take a data-centric approach and provide an integrative framework that blends qualitative and quantitative descriptions of multilabel data sources. By combining lattice theory, in the form of formal concept analysis, and entropy triangles, obtained from information theory, we explain from first principles the fundamental issues of multilabel datasets such as the dependencies of the labels, their imbalances, or the effects of the presence of hapaxes. This allows us to provide guidelines for resampling and new data collection and their relationship with broad modelling approaches. We have empirically validated our framework using 56 open datasets, challenging previous characterizations that prove that our formalization brings useful insights into the task of multilabel classification. Further work will consider the extension of this formalization to understand the relationship between the data sources, the classification methods, and ways to assess their performance.https://www.mdpi.com/2227-7390/12/2/346multilabel classificationmultilabel datasetsinformation sourcesformal concept analysisentropy balancesmeta-analysis |
| spellingShingle | Francisco J. Valverde-Albacete Carmen Peláez-Moreno A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets Mathematics multilabel classification multilabel datasets information sources formal concept analysis entropy balances meta-analysis |
| title | A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets |
| title_full | A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets |
| title_fullStr | A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets |
| title_full_unstemmed | A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets |
| title_short | A Formalization of Multilabel Classification in Terms of Lattice Theory and Information Theory: Concerning Datasets |
| title_sort | formalization of multilabel classification in terms of lattice theory and information theory concerning datasets |
| topic | multilabel classification multilabel datasets information sources formal concept analysis entropy balances meta-analysis |
| url | https://www.mdpi.com/2227-7390/12/2/346 |
| work_keys_str_mv | AT franciscojvalverdealbacete aformalizationofmultilabelclassificationintermsoflatticetheoryandinformationtheoryconcerningdatasets AT carmenpelaezmoreno aformalizationofmultilabelclassificationintermsoflatticetheoryandinformationtheoryconcerningdatasets AT franciscojvalverdealbacete formalizationofmultilabelclassificationintermsoflatticetheoryandinformationtheoryconcerningdatasets AT carmenpelaezmoreno formalizationofmultilabelclassificationintermsoflatticetheoryandinformationtheoryconcerningdatasets |