Automated Multi-Label Classification
Overview
This project presents GA-Auto-MLC and Auto-MEKA_GGP, two methods that automatically select and configure multi-label classification algorithms based on the MEKA tool. If you use this project in your work, please cite the following papers:
- GA-Auto-MLC
- A. G. C. de Sá, G. L. Pappa, and A. A. Freitas. Towards a method for automatically selecting and configuring multi-label classification algorithms. In Proceedings of the Genetic and Evolutionary Computation Conference Companion , pp. 1125–1132, 2017. [ PDF ] [ ACM ]
- Auto-MEKA_GGP
- A. G. C. de Sá, G. L. Pappa, and A. A. Freitas. Automated Selection and Configuration of Multi-Label Classification Algorithms with Grammar-based Genetic Programming. In Proceedings of the International Conference on Parallel Problem Solving from Nature (PPSN), 2018. [To appear.]
- Note that Auto-MEKA_GGP uses the EpochX framework to implement the Grammar-based Genetic Programming (GGP). The framework can be found in the link: EpochX
Documentation
The description of the multi-label classification (MLC) search is available in the following link: MLC Search Space
License
See LICENSE file.
Support
Any questions or comments should be directed to Alex de Sá (alexgcsa@dcc.ufmg.br). You can also create an issue on the repository.
We are still working on a tutorial to turn Auto-MEKA_GGP and GA-Auto-MLC easy methods to use.