HPML: HYBRID PARTITIONS FOR MULTI-LABEL CLASSIFICATION

This repository holds all material developed during my PhD (2019-2023).

PHD MANUSCRIPT

- Qualification

a) Download Manuscript - PtBr

b) Download Presentation - PtBr

c) Video on Youtube - PtBr

- Defense

a) Download Manuscript - PtBr

b) Download Presentation - PtBr

c) Video on Youtube - PtBr

PUBLISHED PAPERS

BRACIS 2023

• Download paper - EnUs
• Download presentation - EnUs
• Video on Youtube - EnUs

IJCNN 2021

• Download paper - EnUs
• Download presentation - EnUs
• Video on Youtube - EnUs

EPPC 2020

• Download paper - PtBr
• Download presentation - PtBr
• Video on Youtube - PtBr

SUBMITTED PAPERS

• Elaine Cecília Gatto, Mauri Ferrandin, Ricardo Cerri. Multi-Label Classification with Label Clusters. Knowledge and Information Systems. 07 July 2023, PREPRINT (Version 1) available at Research Square [https://doi.org/10.21203/rs.3.rs-3133411/v1]

ONGOING PAPERS

• Elaine Cecília Gatto, Felipe Kenji Nakano, Jesse Read, Mauri Ferrandin, Ricardo Cerri and Celine Vens. Label Cluster Chains for Multi-Label Classification.
• Elaine Cecília Gatto, Alan Demétrius Baria Valejo, Mauri Ferrandin, Ricardo Cerri. Community Detection for Multi-Label Classification. Applied Soft Computing. Extended Paper from BRACIS.

PAPERS REPOSITORIES

Experiment 1: HPML.A.c / IJCNN-2021

Experiment 2: EXAUSTIVE-ORACLE / JOURNAL-KAIS-2024

Experiment 3: COMMUNITIES / BRACIS-2023

Experiment 3: COMMUNITIES / JOURNAL-ASOC-2024

Experiment 4: CHAINS / JOURNAL-?-?

ALL RESULTS FOR ALL EXPERIMENTS

Experiment 1: HPML.A.c

Experiment 2: Exhaustive-Oracle

Experiment 3: Communities

Experiment 4: Chains

PLOT HYBRID PARTITIONS

A code to plot the hybrid partitions. You can use this code to plot any label partition. It is very useful to visualize the label clusters, which differ from the instance clusters. Plot HPML

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

[email protected]

Links

| Site | Post-Graduate Program in Computer Science | Computer Department | Biomal | CNPQ | Ku Leuven | Embarcados | Read Prensa | Linkedin Company | Linkedin Profile | Instagram | Facebook | Twitter | Twitch | Youtube |