Hierarchical Multi-Label Software and Datasets
Genetic Algorithms
Here you can find the source code for the method HMC-GA, as described by Cerri et al., 2019. The datasets used in the paper can be downloaded here, from the Declarative Languages and Artificial Intelligence group (DTAI - Katholieke Universiteit Leuven), Belgium. You can also use the Mulan datasets for multi-label flat classification. HMC-GA suports both multi-label tasks (hierarchical and non-hierarchical).
Hierarchical Multi-Label Classification with a Genetic Algorithm - HMC-GA
HMC-GA source code from Cerri et al., 2019
Example of configuration file to be used with HMC-GA for HMC classification (Funcat or GO)
Example of configuration file to be used with HMC-GA for multi-label flat classification
Neural Networks
Here you can find protein function prediction datasets for hierarchical multi-label classification with neural networks. The hierarchies of these datasets were formated to be used with the neural network-based method HMC-LMLP, as described by Cerri et al., 2015 and Cerri et al., 2016. The original datasets can be downloaded here, from the Declarative Languages and Artificial Intelligence group (DTAI - Katholieke Universiteit Leuven), Belgium.
Funcat Datasets (Tree Structure) to be used with HMC-LMLP
Cellcycle | Church | Derisi | Eisen | Gasch1 | Gasch2 | Pheno | Spo | Expr | Seq
Gene Ontology Datasets (DAG Structure) to be used with HMC-LMLP
Cellcycle | Church | Derisi | Eisen | Gasch1 | Gasch2 | Pheno | Spo | Expr | Seq
Hierarchical Multi-Label Classification with Local Multi-Layer Perceptron - HMC-LMLP
HMC-LMLP-Predicted source code from Cerri et al., 2016
Example of configuration file to be used with HMC-LMLP-Predicted (Funcat)
Example of configuration file to be used with HMC-LMLP-Predicted (Gene Ontology)