Dr Heitor Murilo Gomes
Senior Research Fellow & Head of the MOA Lab, AI Institute
Qualifications: BSc UTP MSc PhD PUCPR
Contact Details
Email: hgomes@waikato.ac.nz
Room: FG.2.03
Phone: +64 7 838 4104
Website: http://www.heitorgomes.com
About Heitor Murilo
Heitor is a Senior Research Fellow at the University of Waikato in the machine learning group. He is especially interested in advancing the field of machine learning for evolving data streams, both for fundamental and applied research.
Besides his academic background, Heitor has also worked on several companies, funded a startup, and provided consultancy services on applied machine learning problems.
Research Interests
Heitor's goal is to expand our current knowledge in the field of machine learning for continuous flows of data. This involves the proposal of novel algorithms and theory.
He started working on problems related to machine learning applied to data stream around 2010, and he has continued ever since. In 2019, along with his co-authors, he published a paper on the current state of the field on the ACM SIGKDD Explorations Newsletter.
In 2021, Heitor has been appointed as the head of the MOA Lab at the University of Waikato.
Selected publications
A Survey on Ensemble Learning for Data Stream Classification
H M Gomes, J P Barddal, F Enembreck, A Bifet. ACM Computing Surveys 50, 2, Article 23, 2017. DOI: https://doi.org/10.1145/3054925
This paper contains the most up to date and comprehensive survey about ensemble learning for data streams. Access Paper
Adaptive random forests for evolving data stream classification
H M Gomes, A Bifet, J Read, J P Barddal, F Enembreck, B Pfahringer, G Holmes, T Abdessalem. ACM Machine Learning, Springer, 2017. DOI: https://doi.org/10.1007/s10994-017-5642-8
This paper presents an efficient version of the classical Random Forests algorithm for evolving data streams, namely the Adaptive Random Forest (ARF) algorithm. Access Paper
Streaming Random Patches for Evolving Data Stream Classification
H M Gomes, J Read, A Bifet. IEEE International Conference on Data Mining (ICDM), 2019. DOI: https://doi.org/10.1109/ICDM.2019.00034
The Streaming Random Patches (SRP) algorithm outperforms the current state-of-the-art ensemble methods for evolving data stream classification. Access Paper
Machine learning for streaming data: state of the art, challenges, and opportunities
H M Gomes, J Read, A Bifet, J P Barddal, J Gama. SIGKDD Explorations Newsletter, ACM , 2019. DOI: https://doi.org/10.1145/3373464.3373470
In this work, we focus on elucidating the connections among the current state-of-the-art on related fields; and clarifying open challenges in both academia and industry. Access Paper
Service
Reviewer and/or PC member: Journals
Neurocomputing (Elsevier) - 2016, 2017; Transactions on Big Data (IEEE) - 2017; International Journal of Data Science and Analytics (Springer) - 2017; Information Systems - 2017; Data Mining and Knowledge Discovery - 2018, 2019, 2020; TKDE - 2018, 2019, 2020; IEEE Access - 2019; Journal of Machine Learning (JMLR) - 2019, 2020; Knowledge and Information Systems (KAIS) - 2019;
Reviewer and/or PC member: Conferences
SIGKDD - 2019, 2020, 2021; ECML - 2019, 2020, 2021; PAKDD - 2018, 2019, 2020, 2021; IJCAI - 2018, 2019, 2020, 2021; AAAI - 2019, 2020, 2021, 2022; ICDM - 2017; IEEE International Conference on Big Data - 2017; NeurIPS - 2020, 2021; ESANN 2019; ACM SAC - 2017, 2018, 2019, 2020;
Editor, Chair, Co-chair or Organizer
- Online Experience/Virtual Chair. IEEE International Conference on Data Mining (ICDM) 2021. Auckland, New Zealand. Website.
- Editor of a Special issue of the "Frontiers in Big Data, section Data Mining and Management" journal, 2021.
- ACM Symposium on Applied Computing Data Streams Track. In conjunction with ACM Symposium on Applied Computing: The 36th Annual ACM Symposium on Applied Computing 2021. Gwangju, Korea. Website.
- Second International Workshop on Energy Efficient Scalable Data Mining and Machine Learning. Co-located with ECML PKDD 2019. Würzburg, Germany. Website.
Funding
- Recipient of the University of Waikato Strategic fund for research (Jan-2020 - Dec-2020).
Grants, Awards and Honors
- Added to the Honor Roll of Outstanding Reviewers PAKDD 2020. Website
- PhD Grants. From September 2014 until March 2017 I was recipient of a scholarship from CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) and from Fundação Araucária.
- Best Paper Award Information Systems (ACM SAC 2014). Best paper award for the paper entitled SAE2: Advances on the Social Adaptive Ensemble Classifier for Data Streams.
Publication Statistics
- 36 Full Papers in international conferences
- 16 Papers published in international journals
- 1 Book chapter
- Citations: 1080*; h-index: 15*; i10-index: 20*
All publications: http://heitorgomes.com/publications/
* Statistics according to Google Scholar as of September 2021.
Research Supervised
I've supervised several undergraduate and master level students in different institutions. Currently, I supervise three PhD students at UoW.
If you are interested in working with me, please check the list of research topics that I am currently working on. If you are unsure, you can send me an email to start a conversation.
Teaching Commitments
Recent Publications
Manapragada, C., Gomes, H. M., Salehi, M., Bifet, A., & Webb, G. I. (2022). An eager splitting strategy for online decision trees in ensembles. Data Mining and Knowledge Discovery. doi:10.1007/s10618-021-00816-x
Montiel, J., Halford, M., Mastelini, S. M., Bolmier, G., Sourty, R., Vaysse, R., . . . Bifet, A. (2021). River: Machine learning for streaming data in Python. Journal of Machine Learning Research, 22(10), 1-8. Retrieved from http://jmlr.org/papers/v22/20-1380.html Open Access version: https://hdl.handle.net/10289/14402
Gomes, H., Read, J., Bifet, A., & Durrant, R. J. (2021). Learning from evolving data streams through ensembles of random patches. Knowledge and Information Systems, 63(7), 1597-1625. doi:10.1007/s10115-021-01579-z
Cassales, G., Gomes, H., Bifet, A., Pfahringer, B., & Senger, H. (2021). Improving the performance of bagging ensembles for data streams through mini-batching. Information Sciences, 580, 260-282. doi:10.1016/j.ins.2021.08.085