https://github.com/cissagatto/hpml-chains
This code is a part of my doctoral research at PPG-CC/DC/UFSCar in colaboration with Ku Leuven in Belgium.
https://github.com/cissagatto/hpml-chains
classifier-chains ensemble java label-correlations label-space-partitioning machine-learning multi-label-classification multi-label-clustering multi-label-partition multi-label-partitioning r-language
Last synced: 27 days ago
JSON representation
This code is a part of my doctoral research at PPG-CC/DC/UFSCar in colaboration with Ku Leuven in Belgium.
- Host: GitHub
- URL: https://github.com/cissagatto/hpml-chains
- Owner: cissagatto
- License: gpl-3.0
- Created: 2022-12-27T15:10:52.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2024-09-04T15:37:23.000Z (almost 2 years ago)
- Last Synced: 2024-09-05T21:49:07.512Z (almost 2 years ago)
- Topics: classifier-chains, ensemble, java, label-correlations, label-space-partitioning, machine-learning, multi-label-classification, multi-label-clustering, multi-label-partition, multi-label-partitioning, r-language
- Homepage:
- Size: 2.23 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# HPML Experiment 4: Chains
This is the 4º experiment for HPML which is based on Classifier Chains.
## Enviroments to run experiments
Click here to download
### Conda
You can run this experiment in Conda Environment. The name is "AmbienteTeste". To be able to use this env you must first install conda in your computer or server and then install the environment using the following command: *conda env create --file AmbienteTeste.yaml*
### Singularity
You can also run this experiment in a singularity container. Download the recipes and follow this tutorial (in portuguese): https://prensa.li/@cissa.gatto/tutorial-como-criar-um-container-singularity-para-executar-scripts-r-com-java-e-rclone
## Code
### Step 1
Pre-processing. [10-Fold Cross Validation](https://github.com/cissagatto/CrossValidationMultiLabel)
### Step 2 and 3
Modeling the correlations between the labels and choosing the best dendrogram to generate the hybrid partitions. [Code](https://github.com/cissagatto/jaccard)
### Step 4, 5 and 6
Building and validating all generated hybrid partitions with the silhouette coefficient. The hybrid partition with the highiest silhouet coefficient is chosen to be tested. [Code](https://github.com/cissagatto/Best-Partition-Silhouette)
### Step 7
Building and testing the best chosen hybrid partition.
[HPML.D](https://github.com/cissagatto/HPML.D.padrao)
[HPML.D.CI](https://github.com/cissagatto/HPML.D.CI)
[HPML.D.CE](https://github.com/cissagatto/HPML.D.CE)
[HPML.D.CEI](https://github.com/cissagatto/HPML.D.CEI)
## Global Partitions
[Code](https://github.com/cissagatto/Global-Partitions)
## Local Partitions
[Code](https://github.com/cissagatto/Local-Partitions)
## Download Results
- [Original Multi-Label Datasets](https://cometa.ujaen.es/datasets/)
- [10-Fold Cross Validation Multi-Label Datasets](https://www.4shared.com/s/dYpGZWzjQ)
- [Jaccard Best Dendrograms](https://www.4shared.com/folder/wVsBXIT5/1-Jaccard-Best-Dendrograms.html)
- [Best Partitions from Silhouette Coefficient](https://www.4shared.com/folder/ucwVLJIg/2-Best-Partitions-Silhouette.html)
- Test
## Acknowledgment
- This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.
- This study was financed in part by the Conselho Nacional de Desenvolvimento Científico e Tecnológico - Brasil (CNPQ) - Process number 200371/2022-3.
- The authors also thank the Brazilian research agencies FAPESP financial support.
# Contact
elainececiliagatto@gmail.com
## Links
| [Site](https://sites.google.com/view/professor-cissa-gatto) | [Post-Graduate Program in Computer Science](http://ppgcc.dc.ufscar.br/pt-br) | [Computer Department](https://site.dc.ufscar.br/) | [Biomal](http://www.biomal.ufscar.br/) | [CNPQ](https://www.gov.br/cnpq/pt-br) | [Ku Leuven](https://kulak.kuleuven.be/) | [Embarcados](https://www.embarcados.com.br/author/cissa/) | [Read Prensa](https://prensa.li/@cissa.gatto/) | [Linkedin Company](https://www.linkedin.com/company/27241216) | [Linkedin Profile](https://www.linkedin.com/in/elainececiliagatto/) | [Instagram](https://www.instagram.com/cissagatto) | [Facebook](https://www.facebook.com/cissagatto) | [Twitter](https://twitter.com/cissagatto) | [Twitch](https://www.twitch.tv/cissagatto) | [Youtube](https://www.youtube.com/CissaGatto) |