Classifier chains for multi-label classification

The widely known binary relevance method for multi-label classification, which considers each label as an independent binary problem, has often been overlooked in the literature due to the perceived inadequacy of not directly modelling label correlations. Most current methods invest considerable com...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Machine learning Ročník 85; číslo 3; s. 333 - 359
Hlavní autoři:	Read, Jesse, Pfahringer, Bernhard, Holmes, Geoff, Frank, Eibe
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Boston Springer US 01.12.2011 Springer Springer Nature B.V
Témata:	Acceptability Algorithmics. Computability. Computer arithmetics Applied sciences Artificial Intelligence Chaining Classification Classifiers Complexity Computer Science Computer science; control theory; systems Computer systems and distributed systems. User interface Control Correlation Data processing. List processing. Character string processing Exact sciences and technology Information systems. Data bases Labels Mathematical models Mathematical problems Mechatronics Memory organisation. Data processing Natural Language Processing (NLP) Robotics Simulation and Modeling Software Theoretical computing Problem transformation Scalable methods Ensemble methods Multi-label classification High performance Data analysis Automatic classification Scalability Competitiveness Very large databases Computational complexity Modeling Aggregate model Relevance Model matching Supervised learning Metric Time complexity
ISSN:	0885-6125, 1573-0565
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	The widely known binary relevance method for multi-label classification, which considers each label as an independent binary problem, has often been overlooked in the literature due to the perceived inadequacy of not directly modelling label correlations. Most current methods invest considerable complexity to model interdependencies between labels. This paper shows that binary relevance-based methods have much to offer, and that high predictive performance can be obtained without impeding scalability to large datasets. We exemplify this with a novel classifier chains method that can model label correlations while maintaining acceptable computational complexity. We extend this approach further in an ensemble framework. An extensive empirical evaluation covers a broad range of multi-label datasets with a variety of evaluation metrics. The results illustrate the competitiveness of the chaining method against related and state-of-the-art methods, both in terms of predictive performance and time complexity.
Bibliografie:	SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 14 ObjectType-Article-1 ObjectType-Feature-2 content type line 23
ISSN:	0885-6125 1573-0565
DOI:	10.1007/s10994-011-5256-5