Cookies
Usamos cookies para melhorar nosso site e a sua experiência. Ao continuar a navegar no site, você aceita a nossa política de cookies. Ver mais
Aceitar Rejeitar
  • Menu
Sobre
Download foto HD

Sobre

João Gama é Professor Associado da Faculdade de Economia da Universidade do Porto. É investigador e vice-diretor do LIAAD, INESC TEC. Concluiu o doutoramento na Universidade do Porto, em 2000. É Sénior member do IEEE. Trabalhou em vários projetos nacionais e europeus sobre sistemas de aprendizagem incremental e adaptativo, descoberta de conhecimento em tempo real, e aprendizagem de dados massivos e estruturados. Foi PC chair no ECML2005, DS2009, ADMA2009, IDA '2011 e ECML / PKDD'2015. Foi track chair ACM SAC de 2007 a 2018. Organizou uma série de Workshops sobre Descoberta de Conhecimento de fluxos de dados no ECMLPKDD, ICML, e no ACM SIGKDD. É autor de vários livros em Data Mining e autoria de uma monografia sobre Descoberta de Conhecimento a partir de fluxos de Dados. É autor de mais de 250 papéis peer-reviewed em áreas relacionadas com a aprendizagem automática, aprendizagem de dados em tempo real e fluxos de dados. É membro do conselho editorial de revistas internacionais ML, DMKD, TKDE, IDA, NGC e KAIS. Supervisionou mais de 15 estudantes de doutoramento e 50 alunos de mestrado.

Tópicos
de interesse
Detalhes

Detalhes

  • Nome

    João Gama
  • Cluster

    Informática
  • Cargo

    Investigador Coordenador
  • Desde

    01 abril 2009
018
Publicações

2021

Statistically robust evaluation of stream-based recommender systems

Autores
Vinagre, J; Jorge, AM; Rocha, C; Gama, J;

Publicação
IEEE Transactions on Knowledge and Data Engineering

Abstract

2021

Multi-aspect renewable energy forecasting

Autores
Corizzo, R; Ceci, M; Fanaee T, H; Gama, J;

Publicação
Information Sciences

Abstract
The increasing presence of renewable energy plants has created new challenges such as grid integration, load balancing and energy trading, making it fundamental to provide effective prediction models. Recent approaches in the literature have shown that exploiting spatio-temporal autocorrelation in data coming from multiple plants can lead to better predictions. Although tensor models and techniques are suitable to deal with spatio-temporal data, they have received little attention in the energy domain. In this paper, we propose a new method based on the Tucker tensor decomposition, capable of extracting a new feature space for the learning task. For evaluation purposes, we have investigated the performance of predictive clustering trees with the new feature space, compared to the original feature space, in three renewable energy datasets. The results are favorable for the proposed method, also when compared with state-of-the-art algorithms. © 2020 Elsevier Inc.

2021

Classification and Recommendation With Data Streams

Autores
Veloso, B; Gama, J; Malheiro, B;

Publicação
Encyclopedia of Information Science and Technology, Fifth Edition - Advances in Information Quality and Management

Abstract
Nowadays, with the exponential growth of data stream sources (e.g., Internet of Things [IoT], social networks, crowdsourcing platforms, and personal mobile devices), data stream processing has become indispensable for online classification, recommendation, and evaluation. Its main goal is to maintain dynamic models updated, holding the captured patterns, to make accurate predictions. The foundations of data streams algorithms are incremental processing, in order to reduce the computational resources required to process large quantities of data, and relevance model updating. This article addresses data stream knowledge processing, covering classification, recommendation, and evaluation; describing existing algorithms/techniques; and identifying open challenges.

2021

Forecasting conditional extreme quantiles for wind energy

Autores
Goncalves, C; Cavalcante, L; Brito, M; Bessa, RJ; Gama, J;

Publicação
Electric Power Systems Research

Abstract
Probabilistic forecasting of distribution tails (i.e., quantiles below 0.05 and above 0.95) is challenging for non-parametric approaches since data for extreme events are scarce. A poor forecast of extreme quantiles can have a high impact in various power system decision-aid problems. An alternative approach more robust to data sparsity is extreme value theory (EVT), which uses parametric functions for modelling distribution's tails. In this work, we apply conditional EVT estimators to historical data by directly combining gradient boosting trees with a truncated generalized Pareto distribution. The parametric function parameters are conditioned by covariates such as wind speed or direction from a numerical weather predictions grid. The results for a wind power plant located in Galicia, Spain, show that the proposed method outperforms state-of-the-art methods in terms of quantile score. © 2020 Elsevier B.V.

2021

Hyperparameter self-tuning for data streams

Autores
Veloso, B; Gama, J; Malheiro, B; Vinagre, J;

Publicação
Information Fusion

Abstract

Teses
supervisionadas

2020

Massive Scale Streaming Graphs: Evolving Network Analysis and Mining

Autor
Shazia Tabassum

Instituição
UP-FEUP

2020

Applied Machine Learning Fairness in Business to Consumer Services Industry

Autor
Nuno Filipe Loureiro Paiva

Instituição
UP-FEUP

2020

Visual Similarity with Deep Triplet Quantization: Application to the fashion industry

Autor
Pedro Miguel Vieira Esmeriz Pereira

Instituição
UP-FEP

2020

Detecting fake behavior in online social networks using smartphone data

Autor
Nirbhaya Shaji

Instituição
UP-FCUP

2020

Causal Reasoning in Data

Autor
Ana Rita Dias Nogueira

Instituição
UP-FCUP