Heitor Murilo Gomes

I am a machine learning researcher and developer working at the University of Waikato (New Zealand).

Selected Publications

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 stateof-the-art on related fields; and clarifying open challenges in both academia and industry. Access Paper

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, …, 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 efficent version of the classical Random Forests algorithm for evolving data streams, namely the Adaptive Random Forest (ARF) algorithm. Access Paper

News 2020

  • (December 2020) Students, New MOA version
    • New PhD student starting on February
    • Organized a joint ML seminar with the University of Auckland
    • A new MOA version was released (2020.12). Highlights: new feature analysis tab, new feature importance algorithms, new meta-classifiers for imbalanced classification, kNN for regression. A tutorial about using the novel feature analysis and feature importance tab is available here.
  • (November 2020) Papers
    • Paper accepted at HPCC 2020
    • I am responsible for the international collaborations between UoW and Cardiff University w.r.t AI research
  • (October 2020) Papers, Students and IEEE ICDM 2021
    • Papers accepted at IEEE Big Data 2020 and ROOTS 2020
    • I am the Online Experience/Virtual Chair of ICDM 2021
    • Two master students and one honour student finished this month
  • (September 2020) New website
    • Going through the process of migrating stuff from the old website to the new one.
    • I am co-chair of the DS track at ACM SAC 2021. Read more
    • Attending ECML 2020.
  • (August 2020) The MOA Lab
    • The MOA Lab is open (University of Waikato, FG Link building).
    • One week vacations in the South Island :)
  • (July 2020) scikit-multiflow and papers
    • Paper accepted at Discovery Science 2020.
    • Presented the tutorial entitled ‘Machine learning for data streams in Python with scikit-multi flow’ at IJCNN 2020 with Jacob Montiel, Jesse Read and Albert Bifet.
    • New MOA 20.07 release. Read more.
    • Blog entry about SRP on the MOA website. Read more.
  • (June 2020) COMPX523 wrap-up and papers!
    • Paper accepted at DAWAK 2020.
    • COMPX523 (Data stream mining) 2020 has ended. It was interesting to teach over Zoom, but I prefer to teach in a classroom.
  • (May 2020) Chaired a Session for PAKDD 2020
  • (April 2020) Honour Roll of Outstanding Reviewers for PAKDD 2020
  • (March 2020) COMPX523 and paper
    • Paper accepted at IJCAI 2020
    • COMPX523 Data stream mining paper started at UoW. It was quickly moved to online teaching due to the lockdown.
    • Tutorial 6: Building MOA from the source. Read more.
  • (February 2020) Three papers accepted at IJCNN 2020

  • (January 2020) Distributed ML DS Project kick-off