Matilde Pos-de-Mina Pato

@isel.pt

Department of Electronical Engineering, Telecommunications and Computers (DEETC)
Instituto Superior de Engenharia de Lisboa

Matilde Pos-de-Mina Pato


                        Download Compact PDF  
https://researchid.co/matilde.pato

RESEARCH, TEACHING, or OTHER INTERESTS

Computer Engineering, Information Systems, Health Informatics, Biomedical Engineering
31

Scopus Publications

576

Scholar Citations

8

Scholar h-index

7

Scholar i10-index

Scopus Publications

  • An empirical study on the application of KANs for classification
    Samuel Sampaio Costa, Matilde Pato, Nuno Datia
    Icaai 2024 Conference Proceedings of the 2024 8th International Conference on Advances in Artificial Intelligence, 2025
    Kolmogorov-Arnold Networks (KANs) represent a breakthrough in deep learning, diverging from Multi-Layer Perceptrons (MLPs) by generalizing the Kolmogorov-Arnold representation theorem (KAT) to networks of arbitrary depth and width. This theorem facilitates the decomposition of multivariate functions into constituent one-dimensional elements, with learnable activation functions on weights and the sum operator on nodes. KANs have been shown to exhibit robust performance in function approximation, validated across mathematical, physical, and practical domains such as traffic prediction and medical diagnostics. Our study investigates KANs' efficacy through comprehensive evaluations on OpenML, Kaggle and UCI datasets, with a focus on enhancing Human Activity Recognition systems. They demonstrate high classification performance compared to conventional machine learning approaches and MLPs. These findings underscore KANs' potential as scalable, interpretable tools in modern machine learning applications given their favorable neural scaling laws.
  • Mapping Drug Interactions and Therapeutic Clusters through Knowledge Graph Visualization
    Ana Carolina Pereira, Matilde Pato, Nuno Datia
    Proceedings of the International Conference on Information Visualisation, 2025
    Knowledge graphs (KGs) have emerged as powerful tools for biomedical research, enabling the integration and analysis of heterogeneous data sources. This study explores how a KG, combined with the Neo4j graph database, supports drug analysis and the extraction of biomedical insights. First, data from DailyMed, Purple Book and Orange Book are collected and standardized to ensure interoperability. Second, Named Entity Recognition is applied to address inconsistencies across sources. A hierarchical approach is used to link drugs to Disease Ontology, Orphanet, DrugBank, and ChEBI or active ingredient-based IDs, ensuring data accuracy. The constructed KG facilitates diverse analytical tasks, including the identification of drug-disease associations, longitudinal analysis of drug approval trends, and characterization of common routes of administration. Our results reveal complex interconnections between 561 drugs and 176 diseases, identifying significant regulatory hubs and therapeutic clusters. Temporal analysis demonstrated an acceleration in regulatory activity in recent years, while bipartite network analysis of administration routes revealed predominant delivery methods. Furthermore, graph algorithms provided by Neo4j allow advanced analyses such as finding the shortest paths between drugs based on their regulatory and therapeutic properties, revealing clusters of similar medications, and uncovering candidates for drug repurposing. The findings highlight the potential of KG methodologies in pharmaceutical research, offering a scalable approach to complex biomedical data analysis.
  • Machine learning algorithms to use in the development of maintenance strategies for wind turbines
    J. Sobral, N. Datia, M. Pato, C. Guedes Soares
    Innovations in Renewable Energies Offshore Proceedings of the 6th International Conference on Renewable Energies Offshore Renew 2024, 2025
    This paper addresses machine learning algorithms, which are commonly employed to analyse vast amounts of data collected by condition monitoring systems in wind turbines to identify patterns, anomalies, and trends in the data, enabling predictive maintenance and decision-making. Several possibilities and potential uses of these algorithms in developing predictive maintenance strategies to enhance the reliability and efficiency of wind energy systems are considered. The paper also presents a case study, showing how to deal with data from a Supervisory Control and Data Acquisition system and use it in machine learning algorithms.
  • A Serverless Approach for Resource-constrained Smart Locker Networks
    M. Seleiro, J. Simão, N. Datia, R. Rebelo, M. Pato, A. Serrador, P. Sampaio
    Procedia Computer Science, 2025
    Online retail sales continue to grow, presenting couriers companies with a chance to enhance their market presence. Couriers must keep up with this trend in package processing and delivery. A modern solution is the use of smart lockers. These are equipped with a minimum amount of computing resources, while communicating efficiently with cloud supported services. To address these constraints, developers must consider the system architecture, deployment strategies, and data processing optimisation of the lockers (edge) and the main services (cloud), working alongside stakeholders to acquire the most appropriate hardware for the project. In this paper, we explore the use of a serverless architecture for smart locker. Using stateless functions and workflows, the locker can quickly and effectively react to events from the cloud and its own frontend, while supporting a lightweight serverless framework that provides an easy way to deploy and maintain the continuous software services/functions execution. These serverless frameworks have low resource requirements, allowing developers to utilize less resourceful hardware for the benefit of stakeholders. Additionally, a work-in-progress workflow prototype for the locker functions will also be presented, tested in a resource-constrained environment using faasd as the supporting framework, and compared against a more traditional service-oriented approach – Spring services. Our results show that serverless performed considerably well when compared to service-oriented solutions, with comparable results in terms of both memory usage and response latency.
  • A Unified Communication Architecture for Smart Locker Networks and Mobile Access
    João Silva, Rogério Campos-Rebelo, Nuno Datia, António Serrador, Matilde Pato, José Simão, Pedro Sampaio
    Proceedings 2025 9th International Young Engineers Forum on Electrical and Computer Engineering Yef Ece 2025, 2025
    E-commerce has seen considerable growth, mainly driven by the COVID-19 pandemic. As life returns to normal, challenges with scheduling home deliveries have become more frequent, resulting in a surge in the use of Pick-Up Drop-Off (PUDO) solutions. Smart lockers stand out among these, as they are unattended and allow extended use hours. This work examines various communication strategies between smart-phones, local lockers, and a central server that can support multiple locker hardware and environment configurations. By implementing and testing various approaches, this research aims to identify the advantages and disadvantages of each approach in terms of convenience, reliability, security, compatibility and locker hardware cost. The conclusions of this work help deter-mine, among a set of approaches, which is the most cost-efficient approach, presents advantages and disadvantages depending on environmental conditions, while contributing to the advancement of loT-based cyber-physical systems in the smart locker ecosys-tem. To validate and test these approaches, a mobile application was developed, and the existing backend(local locker and central system) was modified to support the different communication strategies. This work was carried out in collaboration with CTT, a Portuguese postal company.
  • Survey on Recommender Systems for Biomedical Items in Life and Health Sciences
    Matilde Pato, Márcia Barros, Francisco M. Couto
    ACM Computing Surveys, 2024
    The generation of biomedical data is of such magnitude that its retrieval and analysis have posed several challenges. A survey of recommender system (RS) approaches in biomedical fields is provided in this analysis, along with a discussion of existing challenges related to large-scale biomedical information retrieval systems. We collect original studies, identify entities and models, and discuss how knowledge graphs (KGs) can improve results. As a result, most of the papers used model-based collaborative filtering algorithms, most of the available datasets did not follow the standard format < user, item, rating >, and regarding qualitative evaluations of RSs use mainly classification metrics. Finally, we have assembled and coded a unique dataset of 60 papers — Sur-RS4BioT, available for download at DOI:10.34740/kaggle/ds/2346894
  • Enhancing Drug Reviews Insights through Exploratory Data Analysis and Sentiment Analysis
    Ana Sofia Pinto, Matilde Pato, Nuno Datia
    Proceedings of the International Conference on Information Visualisation, 2024
    The increasing volume of user-generated content across various online platforms has created vast datasets in multiple domains, including healthcare. This article explores the significant roles of data visualisation and sentiment analysis within the healthcare sector using the UCI ML Drug Review dataset. Our study highlights the value of exploratory data analysis and sentiment analysis in comprehending patient feedback, enriching insights from the dataset. Data visualisation effectively elucidates the data's distribution and key characteristics, while sentiment analysis, performed using TextBlob and VADER, categorises the emotional tone of patient reviews. Our methodology aims to provide a deeper understanding of patient satisfaction and medication efficacy based on user-generated content.
  • Explainable Feature Ranking Using Interactive Dashboards
    Diogo Amorim, Matilde Pato, Nuno Datia
    Proceedings of the International Conference on Information Visualisation, 2024
    In the dynamic realm of machine learning, achieving transparency and understandability is crucial for fostering trust and facilitating broader adoption. This study presents an enhanced version of the Ensemble Feature Ranking algorithm, tailored to optimize feature selection in machine learning models. This paper proposes the use of an interactive dashboard application, as part of learning environment, designed to provide users with a visually intuitive platform for exploring the algorithm's internal metrics and rankings. The dashboard facilitates a deeper understanding of feature importance and algorithm behaviour, bridging the gap between complex algorithms and user comprehension. By combining advanced algorithmic techniques with a user-centric interface, our approach promotes transparency, accountability and increased user engagement in the explanation of machine learning models.
  • Road Traffic Flow Prediction with Visual Analytics
    Nuno Datia, Matilde P. M. Pato, João Vaz, João Moura Pires
    Studies in Computational Intelligence, 2024
  • How NLP and Visual Analytics Can Improve Asset Management
    Pedro Santos, Matilde P. M. Pato, Nuno Datia, José Sobral
    Studies in Computational Intelligence, 2024
  • Anxolotl, an Anxiety Companion App-Stress Detection
    Nuno Gomes, Matilde Pato, Pedro Santos, Andrè Lourenço, Lourenço Rodrigues
    Human Activity and Behavior Analysis Advances in Computer Vision and Sensors Volume 1, 2024
  • Understanding Portuguese Users of Parcel Locker Services
    Marta Ferreira, Matilde Pato, António Serrador, Rogério Campos-Rebelo, Nuno Datia, Jose Simão, Pedro Sampaio
    Proceedings of the International Conference on Information Visualisation, 2024
  • Recommending Words Using a Bayesian Network
    Pedro Santos, Matilde Pato, Nuno Datia, José Sobral, Noel Leitão, Manuel Ramos Ferreira, Nuno Gomes
    Electronics Switzerland, 2023
  • IRONEDGE: Stream Processing Architecture for Edge Applications
    João Pedro Vitorino, José Simão, Nuno Datia, Matilde Pato
    Algorithms, 2023
  • A Survey on Wearable Sensors for Mental Health Monitoring
    Nuno Gomes, Matilde Pato, André Ribeiro Lourenço, Nuno Datia
    Sensors, 2023
  • Evaluating the Performance of Algorithms in Axillary Microwave Imaging towards Improved Breast Cancer Staging
    Matilde Pato, Ricardo Eleutério, Raquel C. Conceição, Daniela M. Godinho
    Sensors, 2023
  • Data Visualisation on a Mobile App for Real-Time Mental Health Monitoring
    Nuno Gomes, Matilde Pato, André Lourenço, Renato Marcelo, Pedro Santos, Nuno Datia
    Proceedings of the International Conference on Information Visualisation, 2023
  • NLP for Enterprise Asset Management: An Emerging Paradigm
    Pedro Santos, Nuno Datia, Matilde Pato, José Sobral, Nuno Gomes, Noel Leitão, Manuel R. Ferreira
    Proceedings of the International Conference on Information Visualisation, 2023
  • Comparing Word Embeddings through Visualisation
    Pedro Santos, Nuno Datia, Matilde Pato, Jose Sobral
    Proceedings of the International Conference on Information Visualisation, 2022
  • ML Approach to Predict Air Quality Using Sensor and Road Traffic Data
    Nuno Datia, M. P. M. Pato, Ruben Taborda, João Moura Pires
    Studies in Computational Intelligence, 2022
  • Traffic Flow Indicator: Predicting Jams in a City
    Joao Vaz, Nuno Datia, Matilde Pato, Joao Moura Pires
    Proceedings of the International Conference on Information Visualisation, 2022
  • Creating Recommender Systems Datasets in Scientific Fields
    Marcia Barros, Francisco M. Couto, Matilde Pato, Pedro Ruas
    Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2021
  • Exploring air quality using a multiple spatial resolution dashboard-a case study in Lisbon
    Ruben Taborda, Nuno Datia, M.P.M. Pato, Joao Moura Pires
    Proceedings of the International Conference on Information Visualisation, 2020
  • UWB Antenna for Medical Image
    Vitor Cruz, Joao Nuno Matos, Pedro Pinho, M. P. M. Pato
    2018 IEEE Antennas and Propagation Society International Symposium and Usnc Ursi National Radio Science Meeting Apsursi 2018 Proceedings, 2018
  • Integrated electromyography visualization with multi temporal resolution
    Pedro Cardoso, Nuno Datia, M.P.M. Pato
    International Symposium on Medical Information and Communication Technology Ismict, 2017
  • Classification performance of data mining algorithms applied to breast cancer data
    V. Santos, N. Datia, Matilde Pato
    Computational Vision and Medical Image Processing IV Proceedings of Eccomas Thematic Conference on Computational Vision and Medical Image Processing Vipimage 2013, 2014
  • Finite element studies of the mechanical behaviour of the diaphragm in normal and pathological cases
    M. P.M. Pato, N. J.G. Santos, P. Areias, E. B. Pires, M. de Carvalho, S. Pinto, D. S. Lopes
    Computer Methods in Biomechanics and Biomedical Engineering, 2011
  • PBL and engineering: Two sides of its implementation
    Sefi Annual Conference 2011, 2011
  • Active and passive behaviors of soft tissues: Pelvic floor muscles
    M. P. M. Pato, P. Areias
    International Journal for Numerical Methods in Biomedical Engineering, 2010
  • Finite element studies of the deformation of the pelvic floor
    J. A. C. MARTINS, M. P. M. PATO, E. B. PIRES, R. M. N. JORGE, M. PARENTE, T. MASCARENHAS
    Annals of the New York Academy of Sciences, 2007
  • A finite element model of skeletal muscles
    J. A. C. Martins, M. P. M. Pato, E. B. Pires
    Virtual and Physical Prototyping, 2006

RECENT SCHOLAR PUBLICATIONS

  • Why do women pursue a Ph. D. in Computer Science?
    E Ábrahám, M Goulão, MV Janičić, SJ Delany, A Mersni, O Yeremenko, ...
    Journal of Systems and Software, 112586 , 2025
    2025
    Citations: 6
  • Mapping Drug Interactions and Therapeutic Clusters through Knowledge Graph Visualization
    AC Pereira, M Pato, N Datia
    2025 29th International Conference Information Visualisation (IV), 389-395 , 2025
    2025
    Citations: 1
  • A Unified Communication Architecture for Smart Locker Networks and Mobile Access
    J Silva, R Campos-Rebelo, N Datia, A Serrador, M Pato, J Simão, ...
    2025 9th International Young Engineers Forum on Electrical and Computer … , 2025
    2025
  • Optimizing Public Transport with Real-Time Passenger Analytics
    NC João Cravo, Matilde Pato, José Simão
    First International Conference on Transportation Systems (TS2025), 8 , 2025
    2025
  • Knowledge Graphs as Educational Tools in Biomedical Education
    AC Pereira, M Pato, N Datia
    womENcourage™ 2025 Computer Science: a Catalyst for Educational Change , 2025
    2025
  • A Serverless Approach for Resource-constrained Smart Locker Networks
    M Seleiro, J Simão, N Datia, R Rebelo, M Pato, A Serrador, P Sampaio
    Procedia Computer Science 256, 602-609 , 2025
    2025
    Citations: 1
  • Machine learning algorithms to use in the development of maintenance strategies for wind turbines
    J Sobral, N Datia, M Pato, CG Soares
    Innovations in Renewable Energies Offshore, 491-501 , 2024
    2024
    Citations: 1
  • An empirical study on the application of KANs for classification
    SS Costa, M Pato, N Datia
    Proceedings of the 2024 8th international conference on advances in … , 2024
    2024
    Citations: 3
  • Understanding Portuguese Users of Parcel Locker Services
    M Ferreira, M Pato, A Serrador, R Campos-Rebelo, N Datia, J Simão, ...
    2024 28th International Conference Information Visualisation (IV), 45-51 , 2024
    2024
    Citations: 2
  • Explainable Feature Ranking Using Interactive Dashboards
    D Amorim, M Pato, N Datia
    2024 28th International Conference Information Visualisation (IV), 1-6 , 2024
    2024
    Citations: 2
  • Enhancing Drug Reviews Insights through Exploratory Data Analysis and Sentiment Analysis
    AS Pinto, M Pato, N Datia
    2024 28th International Conference Information Visualisation (IV), 190-195 , 2024
    2024
    Citations: 3
  • Road Traffic Flow Prediction with Visual Analytics
    N Datia, MPM Pato, J Vaz, JM Pires
    Artificial Intelligence and Visualization: Advancing Visual Knowledge … , 2024
    2024
    Citations: 1
  • How NLP and Visual Analytics Can Improve Asset Management
    P Santos, MPM Pato, N Datia, J Sobral
    Artificial Intelligence and Visualization: Advancing Visual Knowledge … , 2024
    2024
    Citations: 1
  • Survey on recommender systems for biomedical items in life and health sciences
    M Pato, M Barros, FM Couto
    ACM Computing Surveys 56 (6), 1-32 , 2024
    2024
    Citations: 7
  • Best Practices From Bachelor/Master Studies to Ph. D
    A Mersni, C Oehlhorn, E Abraham, K Boudaoud, L Schmid, M Pato, ...
    EUGAIN, Informatics Europe, COST , 2024
    2024
  • Anxolotl, an Anxiety Companion App–Stress Detection
    N Gomes, M Pato, P Santos, A Lourenço, L Rodrigues
    Human Activity and Behavior Analysis, 147-162 , 2024
    2024
    Citations: 2
  • Recommending words using a bayesian network
    P Santos, M Pato, N Datia, J Sobral, N Leitão, M Ramos Ferreira, ...
    Electronics 12 (10), 2218 , 2023
    2023
    Citations: 2
  • IRONEDGE: Stream processing architecture for edge applications
    JP Vitorino, J Simão, N Datia, M Pato
    Algorithms 16 (2), 123 , 2023
    2023
    Citations: 5
  • Evaluating the performance of algorithms in axillary microwave imaging towards improved breast cancer staging
    M Pato, R Eleutério, RC Conceição, DM Godinho
    Sensors 23 (3), 1496 , 2023
    2023
    Citations: 8
  • A survey on wearable sensors for mental health monitoring
    N Gomes, M Pato, AR Lourenco, N Datia
    Sensors 23 (3), 1330 , 2023
    2023
    Citations: 163

MOST CITED SCHOLAR PUBLICATIONS

  • A survey on wearable sensors for mental health monitoring
    N Gomes, M Pato, AR Lourenco, N Datia
    Sensors 23 (3), 1330 , 2023
    2023
    Citations: 163
  • Finite element studies of the deformation of the pelvic floor
    JAC Martins, MPM Pato, EB Pires, RMN Jorge, M Parente, ...
    Annals of the New York Academy of Sciences 1101 (1), 316-334 , 2007
    2007
    Citations: 149
  • A finite element model of skeletal muscles
    JAC Martins, MPM Pato, EB Pires
    Virtual and Physical Prototyping 1 (3), 159-170 , 2006
    2006
    Citations: 78
  • Ensemble feature ranking applied to medical data
    V Santos, N Datia, MPM Pato
    Procedia Technology 17, 223-230 , 2014
    2014
    Citations: 36
  • Finite element studies of the mechanical behaviour of the diaphragm in normal and pathological cases
    MPM Pato, NJG Santos, P Areias, EB Pires, M De Carvalho, S Pinto, ...
    Computer methods in biomechanics and biomedical engineering 14 (06), 505-513 , 2011
    2011
    Citations: 33
  • Exploring air quality using a multiple spatial resolution dashboard—a case study in lisbon
    R Taborda, N Datia, MPM Pato, JM Pires
    2020 24th International Conference Information Visualisation (IV), 140-145 , 2020
    2020
    Citations: 15
  • Active and passive behaviors of soft tissues: Pelvic floor muscles
    MPM Pato, P Areias
    International Journal for Numerical Methods in Biomedical Engineering 26 (6 … , 2010
    2010
    Citations: 12
  • The ISELthesis LaTeX Template's Manual
    M Pato
    https://github.com/matpato/iselthesis , 2017
    2017
    Citations: 9
  • Evaluating the performance of algorithms in axillary microwave imaging towards improved breast cancer staging
    M Pato, R Eleutério, RC Conceição, DM Godinho
    Sensors 23 (3), 1496 , 2023
    2023
    Citations: 8
  • Survey on recommender systems for biomedical items in life and health sciences
    M Pato, M Barros, FM Couto
    ACM Computing Surveys 56 (6), 1-32 , 2024
    2024
    Citations: 7
  • Why do women pursue a Ph. D. in Computer Science?
    E Ábrahám, M Goulão, MV Janičić, SJ Delany, A Mersni, O Yeremenko, ...
    Journal of Systems and Software, 112586 , 2025
    2025
    Citations: 6
  • Finite element studies of the deformation of the pelvic floor
    JAC Martins, MPM Pato, EB Pires, RMN Jorge, M Parente, ...
    Journal of Biomechanics 39, S627 , 2006
    2006
    Citations: 6
  • IRONEDGE: Stream processing architecture for edge applications
    JP Vitorino, J Simão, N Datia, M Pato
    Algorithms 16 (2), 123 , 2023
    2023
    Citations: 5
  • Comparing word embeddings through visualisation
    P Santos, N Datia, M Pato, J Sobral
    2022 26th International Conference Information Visualisation (IV), 91-97 , 2022
    2022
    Citations: 5
  • Integrated electromyography visualization with multi temporal resolution
    P Cardoso, N Datia, MPM Pato
    2017 11th international symposium on medical information and communication … , 2017
    2017
    Citations: 5
  • A finite element model of skeletal muscle
    B Boubaker, M Pato, E Pires
    Virtual Phys. Prototyp 1 (159-170), 673 , 2006
    2006
    Citations: 5
  • NLP for Enterprise Asset Management: An Emerging Paradigm
    P Santos, N Datia, M Pato, J Sobral, N Gomes, N Leitão, MR Ferreira
    27th International Conference Information Visualisation (IV), 238-243 , 2023
    2023
    Citations: 4
  • An empirical study on the application of KANs for classification
    SS Costa, M Pato, N Datia
    Proceedings of the 2024 8th international conference on advances in … , 2024
    2024
    Citations: 3
  • Enhancing Drug Reviews Insights through Exploratory Data Analysis and Sentiment Analysis
    AS Pinto, M Pato, N Datia
    2024 28th International Conference Information Visualisation (IV), 190-195 , 2024
    2024
    Citations: 3
  • Understanding Portuguese Users of Parcel Locker Services
    M Ferreira, M Pato, A Serrador, R Campos-Rebelo, N Datia, J Simão, ...
    2024 28th International Conference Information Visualisation (IV), 45-51 , 2024
    2024
    Citations: 2