Tsukuba University, Japan, 2020
The Atomium, Brussels, Belgium, 2016.
Juan Gabriel Colonna
Introducing myself: I'm a young and passionate researcher with a Ph.D. in Computer Science, and I hold a Master's degree in the same field, as well as a background in Telecommunication Engineering. Currently, I am a professor at the Institute of Computing at the esteemed Federal University of Amazonas. I bring a wealth of experience in Research and Development (R&D) having worked at Samsung Computer Development Institute of Amazonas (SIDIA). I'm an active member of the Laboratory of Mobile and Pervasive Computing (LCMU) and the Machine Learning Group at the Federal University of Amazonas. My research interests span across Ecological Informatics, Machine Learning, Deep Learning, Signal Processing, Sensor Networks, and Environmental Monitoring Applications in general, with a keen focus on developing innovative solutions to real-world problems.
Link to my CV (portugués version): http://lattes.cnpq.br/9535853909210803 (Spanish version)
Current affiliation: Federal University of Amazonas (UFAM, Icomp), Manaus, Brazil.
Contact: juancolonna@icomp.ufam.edu.br
Idioms: Spanish (mother language), Portuguese (Fluent), English (Intermediate)
GitHub: https://github.com/juancolonna
Google Scholar: https://scholar.google.com.br/citations?user=wQdRmk4AAAAJ&hl=en
Work Experience (R&D experience):
Currently: graduate program coordinator
Currently: Adjunct Professor at Federal University of Amazonas (UFAM)
Past: SIDIA (Samsung, 2018): Software Development and IA specialist.
Past: National University of Río Cuarto (2013)
Academic degrees:
Telecommunications Engineer, Universidad Nacional de Río Cuarto (UNRC), Argentina, 2009-dec.
M.Sc. Computer Science, Universidade Federal do Amazonas (UFAM), Brazil, 2012-feb.
Ph.D. Computer Sciences, Universidade Federal do Amazonas (UFAM), Brazil, 2017-sep. (Thesis PDF portugues version)
Awards and recognitions:
Master Dissertation Award, 1st place in the XX Latin American Computing Contest sponsored by CLEI (Congreso Latinoamericano en Informática). .
"Vamos publicar". Institutional Recognition for the contribution to the Intellectual Production of UFAM.
Best Paper Award in IX Brazilian Symposium on Ubiquitous and Pervasive Computing (SBCUP). ‘Sensor Acústico para Detecção de Desmatamento Ilegal na Floresta Amazônica’. Brazilian Society of Computing.
Additional training courses:
Large Multimodal Model Prompting with Gemini, 2024-september (deeplearning.ia, certificate)
Neural Networks and Deep Learning. Coursera, 2018-jun (deeplearning.ia, certificate)
Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization, 2018-jul (deeplearning.ia, certificate)
Structuring machine learning projects. Coursera, 2018-jul (deeplearning.ia, certificate)
Convolutional Neural Networks. Coursera, 2018-aug (deeplearning.ia, certificate)
Sequence Models. Coursera, 2018-aug (deeplearning.ai, certificate)
Programming skills:
Fluency: Python, C, R, Matlab
Frameworks: Sklearn, Keras, Tensorflow, SciPy
International experience:
Visiting researcher at Tsukuba University, Japan (January-2020), under the supervision of Professor Kazuhiro Fukui.
Visiting researcher at National Unversity of Río Cuarto, Argentina (November-2019), under the supervision of Professor Marcelo Ruiz.
Doctoral Internship at Artificial Intelligence and Decision Support (LIAAD) at INESC TEC, Portugal (2015-2016), under the supervision of Professor João Gama.
Visiting researcher at Federal University of Alagoas, Brazil (October-2014), under the supervision of Professor Osvaldo Rosso.
Teaching experience:
Computer Organization
Computer Architecture
Data Structures and Algorithms using C
Introduction to Python Programming
Data Science with Tradicional Machine Learning (Hands-on course)
Deep Neural Networks (Hands-on course)
Online courses (https://sites.google.com/icomp.ufam.edu.br/oc-icomp/v%C3%ADdeo-aulas)
Journals and book chapters:
2024 - Process mining embeddings: Learning vector representations for Petri nets. Intelligent Systems with Applications (PDF)
2024 - Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy. Journal of the Brazilian Computer Society (PDF)
2023 - Bag of tricks for long-tail visual recognition of animal species in camera-trap images. Ecological Informatics. (PDF)
2023 - Monitoramento Ambiental Não Invasivo Utilizando Dados de Sensores e Técnicas de Aprendizagem de Máquina (PDF, página 24)
2023 - Discriminative Singular Spectrum Classifier with applications on bioacoustic signal recognition. Digital Signal Processing, Volume 133. (PDF)
2022 - How Validation Methodology Influences Human Activity Recognition Mobile Systems. Sensors, MDPI v 22, Article 2360 (PDF)
2022 - Using amino acids co-occurrence matrices and explainability model to investigate patterns in dengue virus proteins. BMC Bioinformatics. v 23, Article 80 (PDF)
2021 - Short Communication: Optimally Solving the Unit-Demand Envy-Free Pricing Problem with Metric Substitutability in Cubic Time. Algorithms 2021, 14(10), 279. MDPI (link, PDF)
2020 - Bioacoustic signal denoising: a review. Artificial Intelligence Review. Springer (link, PDF)
2020 - Estimating ecoacoustic activity in the Amazon rain forest through Information Theory quantifiers. Plos One. PDF
2020 - A Smartphone Lightweight Method for Human Activity Recognition Based on Information Theory. Sensors, MDPI (link)
2020 - Deep hashing for multi-label image retrieval: a survey. Artificial Intelligence Review. Springer (link)
2018 - Unsupervised Selection of the Singular Spectrum Components Based on Information Theory for Bioacoustic Signal Filtering. Digital Signal Processing. Elsevier. (link)
2018 - A comparison of hierarchical multi-output recognition approaches for anuran classification. Machine Learning. Springer. (Article link)
2018 - Feature evaluation for unsupervised bioacoustic signal segmentation of anuran calls. Expert Systems With Applications. Elsevier. (Article link)
2015 - An Incremental Technique for Real-Time Bioacoustic Signal Segmentation. Expert Systems With Applications. Elsevier. (Article link, dataset #2)
2016 - How to Correctly Evaluate an Automatic Bioacoustics Classification Method. Advances in Artificial Intelligence (LNCS). Springer. (link)
2016 - Recognizing Family, Genus, and Species of Anuran Using a Hierarchical Classification Approach. Discovery Science (LNCS). Springer. (Article link, dataset #1)
2016 - Experimental Evaluation on Machine Learning Techniques for Human Activities Recognition in Digital Education Context. Communications in Computer and Information Science . Springer. (pdf)
ArXiv:
2024 - Tasks People Prompt: A Taxonomy of LLM Downstream Tasks in Software Verification and Falsification Approaches (link)
2023 - Catastrophic Forgetting in Deep Learning: A Comprehensive Taxonomy (link)
2023 - Bag of Tricks for Long-Tail Visual Recognition of Animal Species in Camera-Trap Images (link)
2023 - A note on improving the search of optimal prices in envy-free perfect matchings (link)
2022 - New Metrics for Learning Evaluation in Digital Education Platforms (link)
2021 - Discriminative Singular Spectrum Classifier with Applications on Bioacoustic Signal Recognition (link)
2021 - Filtering Empty Camera Trap Images in Embedded Systems (link)
BiorXiv:
2020 - Quantifying ecoacoustic activity in the Amazon rainforest through Information Theory quantifiers (link)
Conferences:
2024 - Enhancing widget recognition for automated Android testing using Computer Vision techniques (WEBMEDIA, PDF)
2024 - Complex-valued Embedding on Grassmann Manifolds for Pattern set Representation (SIBGRAPI, PDF)
2024 - Detection of Elderly Falls in Video Streams Using Skeleton Key Points and Transformer Networks (SIBGRAPI, PDF)
2024 - Recuperação semântica de paisagens sonoras usando banco de dados vetoriais (WCAMA, PDF)
2024 - Moeda Eco: Uma solução inovadora para promover a reciclagem e recompensar ações sustentáveis (WCAMA, PDF)
2024 - A Hierarchical Approach for Extracting and Displaying Entities and Relations from Radiology Medical Reports (SBCAS, PDF)
2023 - Classification of Tropical Disease-carrying Mosquitoes Using Deep Learning and SHAP. (SBCAS, PDF)
2022 - Autenticação de alunos utilizando dinâmica de digitação e redes neurais profundas em ambientes Juiz On-line. (PDF)
2022 - Real Time Detection of Mobile Graphical User Interface Elements Using Convolutional Neural Networks (PDF)
2022 - Q-funcT: A Reinforcement Learning Approach for Automated Black Box Functionality Testing (PDF)
2022 - Investigando a relação entre os aminoácidos de proteínas do vírus da dengue e o desfecho clínico do paciente (PDF)
2021 - Um Método de Detecção de Plágio para Sistemas Juiz On-line baseado no Comportamento dos Alunos (PDF)
2021 - Filtering Empty Camera Trap Images in Embedded Systems (IEEE - CVPR, PDF)
2021 - An Incremental Version of Singular Spectrum Analysis (MACI, PDF)
2021 - RQA análise e classificação de sequências genômicas do vírus da dengue. Mateus Medeiras de Souza, Juan Colonna, Leonardo Souza. Anais do XVIII Congresso Brasileiro de Informática em Saúde e 10º Congresso Brasileiro de Telemedicina e Telessaúde (PDF)
2021 - Classificação de diferentes tipos de câncer de pele, utilizando de redes neurais de convolução. José Cabrera, Juan Colonna, Fagner Cunha. Anais do XVIII Congresso Brasileiro de Informática em Saúde e 10º Congresso Brasileiro de Telemedicina e Telessaúde (PDF)
2021 - Classificação multirrótulo de patologias pulmonares em imagens de radiografias utilizando redes neurais de convolução. Aldemir Silva, Juan Colonna, Fagner Cunha. Anais do XVIII Congresso Brasileiro de Informática em Saúde e 10º Congresso Brasileiro de Telemedicina e Telessaúde (PDF)
2020 - Discriminative Singular Spectrum Analysis for bioacoustic classification. (INTERSPEECH, PDF)
2020 - Autenticação contínua de alunos utilizando biometria comportamental em ambiente Juiz On-line (SBIE, PDF)
2020 - Aprendizagem de Máquina para Classificação de Doenças Respiratórias: Uma Revisão Sistemática. Silva, L.B.; Nogueira, F.N.; Santos, J.M.; Quispe, K.G.M.; Giusti, R.; Colonna, J.G. (pdf)
2019 - SVC-A2C - Actor Critic Algorithm to Improve Smart Vacuum Cleaner. IX Simpósio Brasileiro de Engenharia de Sistemas Computacionais. (Article)
2019 - Previsão de ocupação de anuros usando sensoriamento de variáveis ambientais e modelos Autocodificadores. Anais do XI Simpósio Brasileiro de Computação Ubíqua e Pervasiva e II Workshop Brasileiro de Cidades Inteligentes. XXXIX Congresso da Sociedade Brasileira de Computação. (Article)
2019 - Bioacoustic Complexity Index. VII Congreso de Matemática Aplicada, Computacional e Industrial (MACI). Argentina. (Article, Presentation español)
2017 - Mutual Singular Spectrum Analysis for Bioacoustics Classification. IEEE International Workshop on Machine Learning for Signal Processing (MLSP). (Article, Dataset)
2017 - Sensor Acústico para Detecção de Desmatamento Ilegal na Floresta Amazônica. IX Simpósio Brasileiro de Computação Ubiqua e Pervasiva (SBCUP) . (pdf)
2017 - Detecção de Desmatamento Ilegal na Floresta Amazônica Baseada em Processamento de Áudio. Concurso de Trabalhos de Iniciação Científica (CTIC). (Article)
2017 - Framework para coleta e inferência de estados emocionais de alunos baseado em reconhecimento de expressões faciais. Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação - SBIE). (Article)
2016 - Automatic Classification of Anuran Sounds Using Convolutional Neural Networks. Ninth International C* Conference on Computer Science & Software Engineering . (link)
2016 - A Framework for Chainsaw Detection Using One-Class Kernel and Wireless Acoustic Sensor Networks into the Amazon Rainforest. 15th International Conference on Information Processing in Sensor Networks (IPSN) . (link)
2014 - A Distribute Approach for Classifying Anuran Species Based on Their Calls. 22nd International Conference on Pattern Recognition (ICPR). (Article)
2012 - Feature Subset Selection for Automatically Classifying Anuran Calls Using Sensor Networks. International Joint Conference on Neural Networks (IJCNN) . (Article link)
2012 - Similarity Clustering for Data Fusion in Wireless Sensor Networks Using k-Means. International Joint Conference on Neural Networks (IJCNN).
2012 - Compressão de Vocalizações de Anuros para Classificação de Espécies Usando Redes de Sensores Sem Fio
2012 - Compressive Sensing for Efficiently Collecting Wildlife Sounds with Wireless Sensor Networks. 21st International Conference on Computer Communications and Networks (ICCCN).
2011 - Classificação de Anuros Baseado em Vocalizações para Monitoramento Ambiental Pervasivo. Simpósio Brasileiro de Computação Ubiqua e Pervasiva (SBCUP).
2010 - Algoritmos de Localização Recursivos em Nós Sensores SunSpot. ERIN.
2010 - Mobilidade Urbana em Manaus: Proposta de Otimização de Rotas para o Monotrilho. ERIN.
Presentations:
Estimating ecoacoustic activity in the Amazon rainforest through Information Theory quantifiers. Seminar organized by Laboratório de Computação Científica e Análise Numérica (LACCAN) da Universidade Federal de Alagoas, 2020-10. (pdf)
SEMINFO 2016 - Anuran species recognition using a hierarchical classification approach (pdf)
How to Correctly Evaluate an Automatic Bioacoustics Classification Method. 17th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2016, Salamanca, Spain. (pdf)
Anuran species recognition using a hierarchical classification approach. 19th International Conference on Discovery Science (DS 2016), Bari, Italy. (pdf)
Uma abordagem para monitoramento de anuros baseada em processamento digital de sinais bioacústicos (pdf)
Posters:
A Hierarchical Classification Model for Anuran Calls Recognition. MAESTRA SUMMER SCHOOL ON MINING BIG AND COMPLEX DATA, 04/Sep/2016, Ohrid, Macedonia.(pdf)
2014 - Permutation entropy applied to bioacoustic signal segmentation. XVIII Conference on Nonequilibrium Statistical Mechanics and Nonlinear Physics.
2014 - A Distribute Approach for Classifying Anuran Species Based on Their Calls (pdf)
Mentoring:
Formed Graduate students
2024 - Larissa de Andrade Silva. Explorando variações do canto do Curruíra sob ruído antropogênico: uma abordagem orientada por XAI para análise e interpretação bioacústica. Master Level
2022 - Ronem Matos Lavareda Filho. Student authentication using behavioral biometrics in an online judge environment. Master Level
2022 - Marcos Marreiro Salvatierra. Improvement of Performance in Determining Optimal Envy-Free Pricing. PhD Level.
2021 - Leonardo Rodrigues de Souza. Representation, classification, and interpretation of protein sequences of the dengue virus. Master Level
2019 - Francisco Fagner do Rego Cunha. A study on approaches for out-of-sample evaluation of animal classification models in camera trap images. Master Level (Co-advisor)
Formed Undergraduate students
2024 - Matteo Freitas Reis. Detection of Elderly Falls in Video Streams Using Skeleton Key Points and Transformer Networks.
2023 - Diego Mamede F. Queiroz. Investigando a relação entre os aminoácidos de proteínas do vírus da dengue e o desfecho clínico do paciente.
2023 - Airton Azevedo Franco Martins. ECOLOGGER: Low-Cost Open-Source Device for Ecoacoustic Environmental Monitoring
2022 - Matheus Miranda Matos. Emotion Recognition From Audio Using Deep Learning
2021 - Mateus Medeiros de Souza. RQA analysis and classification of genomic sequences of the dengue virus
2021 - Richard Hada Degaki. Detection of User Interface Elements in Android Applications.
2020 - Lohana Patrícia Soares Viana. A CNN Model Using Bluetooth Low-energy and RSSI for Indoor Cat Detection
2019 - José Alberto Gurgel Cardoso Neto. IoT System Based on YOLOv3 Algorithm for Electricity Management
2019 - Mateus Oliveira da Silva. Reconstruction of High-Frequency Wavelet Coefficients Using Convolutional Neural Networks
2019 - Nabson Paiva Souza da Silva. Prediction Model of Animal Species Distribution Using Probabilistic Neural Networks.
2018 - Willians Cassiano de Freitas Abreu. LSTM Networks With CTC Loss Applied to Gesture-Based Typing Decoding (Co-advisor)
2017 - Waldomiro J. Seabra. Illegal logging prevention with automatic chainsaw detection using acoustic sensors (Co-advisor)
Examination Board Participation (incomplete list)
2017 - Evaluation Committee Member of Carlos Vicente S. Araújo graduate student. Institute of Computing (Icomp), UFAM. Predicting Music Success Based on Users' Comments on Online Social Networks.
2017 - Evaluation Committee Member of Gilvan O. dos Reis graduate student. Institute of Computing (Icomp), UFAM. Utilização dos metadados do Twitter para a análise da repercussão de notícias criminais na cidade de Nova Iorque.