Cookies Policy
We use cookies to improve our site and your experience. By continuing to browse our site you accept our cookie policy. Find out More
Close
  • Menu
About

About

Adjunt professor at Computers Engineering Department, ESTG-Leiria (Polytechnic of Leiria) and reseracher at CRACS.

Holds a PHD in Computer Science by Universidade do Porto; MSc in Informatics, branch of systems and networks, also by Universidade do Porto; Degree in Computers Enginnering by Instituto Superior de Engenharia do Porto (Polytechnic of Porto).

Coordinates a post-graduation on security and digital forensics at Polytechnic of Leiria and is responsible by classes on networking, systems administration, cloud technology and datacenters infrastrucutres.

Main areas of research include immune-inspired algorithms applied to automatic detection of anomalies, ensemble based algorithms for classification and anomaly detection, learning on dynamic systems in a temporal basis.

Previously he was algo ICT project manager and system administrator in companies.

Interest
Topics
Details

Details

  • Name

    Mário João Antunes
  • Cluster

    Computer Science
  • Role

    Project Leader
  • Since

    01st January 2009
Publications

2017

Automatic Documents Counterfeit Classification Using Image Processing and Analysis

Authors
Vieira, R; Antunes, M; Silva, C; Assis, A;

Publication
Pattern Recognition and Image Analysis - 8th Iberian Conference, IbPRIA 2017, Faro, Portugal, June 20-23, 2017, Proceedings

Abstract

2017

Performance Metrics for Model Fusion in Twitter Data Drifts

Authors
Cósta, Joana; Silva, Catarina; Antunes, Mario; Ribeiro, Bernardete;

Publication
Pattern Recognition and Image Analysis - 8th Iberian Conference, IbPRIA 2017, Faro, Portugal, June 20-23, 2017, Proceedings

Abstract

2017

Adaptive learning for dynamic environments: A comparative approach

Authors
Cósta, J; Silva, C; Antunes, M; Ribeiro, B;

Publication
Eng. Appl. of AI

Abstract

2017

Adaptive learning for dynamic environments: A comparative approach

Authors
Costa, J; Silva, C; Antunes, M; Ribeiro, B;

Publication
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE

Abstract
Nowadays most learning problems demand adaptive solutions. Current challenges include temporal data streams, drift and non-stationary scenarios, often with text data, whether in social networks or in business systems. Various efforts have been pursued in machine learning settings to learn in such environments, specially because of their non-trivial nature, since changes occur between the distribution data used to define the model and the current environment. In this work we present the Drift Adaptive Retain Knowledge (DARK) framework to tackle adaptive learning in dynamic environments based on recent and retained knowledge. DARK handles an ensemble of multiple Support Vector Machine (SVM) models that are dynamically weighted and have distinct training window sizes. A comparative study with benchmark solutions in the field, namely the Learn + +.NSE algorithm, is also presented. Experimental results revealed that DARK outperforms Learn + +.NSE with two different base classifiers, an SVM and a Classification and Regression Tree (CART).

2016

Choice of Best Samples for Building Ensembles in Dynamic Environments

Authors
Costa, J; Silva, C; Antunes, M; Ribeiro, B;

Publication
ENGINEERING APPLICATIONS OF NEURAL NETWORKS, EANN 2016

Abstract
Machine learning approaches often focus on optimizing the algorithm rather than assuring that the source data is as rich as possible. However, when it is possible to enhance the input examples to construct models, one should consider it thoroughly. In this work, we propose a technique to define the best set of training examples using dynamic ensembles in text classification scenarios. In dynamic environments, where new data is constantly appearing, old data is usually disregarded, but sometimes some of those disregarded examples may carry substantial information. We propose a method that determines the most relevant examples by analysing their behaviour when defining separating planes or thresholds between classes. Those examples, deemed better than others, are kept for a longer time-window than the rest. Results on a Twitter scenario show that keeping those examples enhances the final classification performance.