Ver o conteúdo principal

Teaching Staff

Biography

João Fonseca is an invited professor and PhD Candidate at NOVA Information Management School. He is working with Prof. Fernando Bação on synthetic data generation, active learning, and data preprocessing methods. Specifically, his research focuses on improving the quality of Land Use/Land Cover classification tasks through the application of these data preprocessing methods. João Fonseca's research is funded by an MIT Portugal PhD Grant (2020 FCT-MPP2030).
In the past, João Fonseca conducted research on Land Use/Land Cover classification methods to automatically update LULC maps of the Portuguese mainland. His work included the development of pipelines to systematize the preprocessing of Sentinel-2 satellite imagery for any given period. He also developed and deployed different types of algorithms for various tasks, such as data filtering, dimensionality reduction, feature extraction, and classification.

Scientific Publications

Fonseca, J., & Bação, F. (2022)

Research Trends and Applications of Data Augmentation Algorithms. (pp. 1-23). Cornell University (ArXiv). https://doi.org/10.48550/arXiv.2207.08817

Fonseca, J., Douzas, G., & Bacao, F. (2021)

Improving imbalanced land cover classification with k-means smote: Detecting and oversampling distinctive minority spectral signatures. Information (Switzerland), 12(7), 1-20. [266]. https://doi.org/10.3390/info12070266

Fonseca, J., Douzas, G., & Bacao, F. (2021)

Increasing the effectiveness of active learning: Introducing artificial data generation in active learning for land use/land cover classification. Remote Sensing, 13(13), 1-20. [2619]. https://doi.org/10.3390/rs13132619

Douzas, G., Bacao, F., Fonseca, J., & Khudinyan, M. (2019)

Imbalanced learning in land cover classification: Improving minority classes' prediction accuracy using the geometric SMOTE algorithm. Remote Sensing, 11(24), [3040]. https://doi.org/10.3390/rs11243040