Rethinking Computer Engineering in the Age of Generative Artificial Intelligence

On the 14th of July, the Department of Informatics Engineering (DEI) at FEUP will host another DEI Open Day. This year’s event will focus on ‘Rethinking Computer Engineering in the Age of Generative Artificial Intelligence‘.

The emergence of generative artificial intelligence, particularly large language models, is transforming the way we think about, teach and practise informatics engineering. Activities traditionally associated with human reasoning, such as writing, synthesising knowledge and programming, are now being shared with systems that learn, reason and create.

This new context raises fundamental questions: Which skills should we prioritise in the training of computer engineers? What should still be taught? How should higher education institutions prepare students for a rapidly changing job market?

In his keynote speech, Luís Garrido Marques (PBS) will address these questions, and in the roundtable discussion, Carlos Baquero (DEI/FEUP) will bring together Luís Wolff Barbosa (L.EIC), Pedro Fortuna (Jscrambler), Helena Leite (Cleva) and Miguel Sozinho Ramalho (Bellingcat) will explore these questions and others, offering academic, student and business perspectives on the major transformations taking place in today’s world.

During the afternoon, company representatives will have the opportunity to hold individual sessions with DEI faculty members to learn more about institutional collaboration possibilities, particularly in the context of Master’s theses, the Integrative Project of the Bachelor’s Degree in Informatics and Computing Engineering (L.EIC), the Project Management Laboratory (LGP) and programmes such as FEUP Prime and DoRPE. Representatives will also be able to explore partnerships with student groups associated with DEI.

The annual DEI Open Day showcases the teaching, research and innovation activities of the DEI, with the aim of strengthening ties with companies, institutions and society in general and promoting impactful collaborations.

More information about the event can be found at the event´s website.

DEI Talks | “Immersive Media and the Colombian Armed Conflict: Rethinking Journalism Through 360º Storytelling” by Andrés Lotero

The talk “Immersive Media and the Colombian Armed Conflict: Rethinking Journalism Through 360º Storytelling” will be presented July the 10th, at 16:30, na in room B010, and will be moderated by António Baía Reis (Assistant Professor and Marie Skłodowska-Curie Postdoctoral Fellow – Department of Sociology and Communication, University of Salamanca, Spain).

Abstract:

This talk presents research on the use of immersive media, specifically 360º video technology, as a journalistic tool for covering the Colombian armed conflict and promoting narratives of peacebuilding. Drawing on a qualitative multiple-case study, the speaker explores four immersive projects that document post-conflict experiences from the perspectives of victims and former combatants.The presentation will examine the motivations behind the adoption of immersive technologies in these contexts and critically reflect on their potential, limitations, and ethical implications. Particular attention will be given to the emotional and narrative impact of immersive storytelling, as well as concerns regarding image manipulation and journalistic independence. By highlighting good practices and identifying key challenges, this talk contributes to a broader understanding of the role immersive media can play in conflict reporting and peace communication, offering valuable insights for journalists, media professionals, and scholars interested in emerging storytelling formats. In an increasingly polarized and crisis-driven world, this research invites reflection on how immersive journalism can foster deeper engagement with complex realities.

About the Speaker:

Andrés David Castro Lotero is a teacher of Spanish language with a focus on Culture and Economics and a Ph.D. candidate in Communication Sciences at the University of Passau since 2019. He has previously studied Communication and Journalism in Colombia and did a master’s degree in Development Studies at the University of Passau. His research interest is mainly the use of new media in the development of vulnerable communities and peacebuilding, especially in Latin American contexts.

PhD Defence in Digital Media: “Integration of models for linked data in cultural heritage and contributions to the FAIR principles”

Candidate:
Inês Dias Koch

Date, Time and Location
1st of July 2025, 14:30, Sala de Atos da Faculdade de Engenharia da Universidade do Porto

Title:
“Integration of models for linked data in cultural heritage and contributions to the FAIR principles”

President of the Jury:
João Carlos Pascoal Faria (PhD), Full Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

Members:
Maja Žumer (PhD), Full Professor, Department of Library and Information Science of the University of Ljubljana, Slovenia;

María Poveda Villalón (PhD), Associate Professor, Departament of Artificial Intelligence of the Technical University of Madrid, Spain;

José Luís Brinquete Borbinha (PhD), Full Professor, Department of Computer Science and Engineering, Instituto Superior Técnico da Universidade de Lisboa;

Pedro Manuel Rangel Santos Henriques (PhD), Full Professor, Department of Informatics, Escola de Engenharia da Universidade do Minho;

Carla Alexandra Teixeira Lopes (PhD), Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto (Supervisor);

Mariana Curado Malta (PhD), Assistant Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

The thesis was co-supervised by Maria Cristina de Carvalho Alves Ribeiro (PhD), Retired Associate Professor in the Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

Abstract:

The various areas of Cultural Heritage, such as archives, museums, and libraries, have invested in tech-
nological development and innovation to make their resources available to users more efficiently and completely. To this end, the description of these resources is essential so that they are explained in
terms of their context and content, as well as to facilitate their intelligibility and accessibility. In this
sense, each area has begun to develop its own models and standards for describing the cultural objects
it deals with. This has made these standards sector-specific and only able to fulfil the information needs
within the area of knowledge they were developed, exploring only the information described within their
domain. As a result, linking resources from different information sources is challenging.
With the need to make the standards and models more interoperable, linked data models emerged in
Cultural Heritage. These models make it possible to link the various concepts from the different heritage
areas efficiently and effectively, considering the Semantic Web’s characteristics.
In Portugal, the Portuguese National Archives felt the need to develop a linked data model to describe
their cultural objects, which led to the creation of the EPISA Project, the project from which this research
emerged. Thus, this work aims to develop a linked data model to describe archival records, as well as to
connect them with other heritage domains, integrating them with existing linked data models, promoting
the access and reuse of data from heritage institutions based on the specialised description associated
with the cultural objects of these institutions. Additionally, it aims to link existing data models to data from other sources available on the Web, such as Wikidata and DBpedia.
We carry out a study that includes existing data models in Cultural Heritage, such as CIDOC CRM
in museums, RiC-CM in archives, and LRMoo in libraries, along with models that have emerged within
Web projects, such as DBpedia and Wikidata. By describing archival objects, as well as creating and
exploring relationships between other data models, this study identifies common characteristics and
principles, as well as the distinctive aspects of each area. Furthermore, it identifies the possibility of linking elements of the various models, ensuring that the models can be adapted to applications without losing the richness of the conceptualisation carried out in each of the domains.
In a context in which the Web promotes the explicitness of data semantics through the Semantic
Web and provides tools to represent it, it is necessary, on the one hand, to create links between models from different communities and, on the other, to adjust the complexity of each model to each application according to its specific requirements. The FAIR Principles (Findable, Accessible, Interoperable, Reusable) were therefore used as one of the sources for the requirements that data and metadata must fulfil to have a modular structure. We bring together a collection of use cases linked to archives users, including profiles ranging fromcollection managers to heritage promoters and informal users. In addition, we compile and evaluate a set of data modelling experiences using different models.
This work resulted in ArchOnto, a modular ontology that describes archive records. It was developed
considering existing archive standards and validated by experts in the field, specifically archivists from the Portuguese National Archives. ArchOnto is based on CIDOC CRM, combined with four other
specific ontologies also developed in this work.
The development of ArchOnto led to the creation of a prototype platform designed to explore and
manipulate archive records. Additionally, it offers the potential to apply this ontology to other domains, specifically to the representation of cinematographic records.

Keywords: Cultural Heritage; Linked Open Data; Data Integration; Semantic Web; FAIR Principles; Digital Humanities.

Research “Made in FEUP” exhibited at DCE 2025

The DCE – Doctoral Congress in Engineering, is back for its sixth edition, showcasing hundreds of research projects currently underway in various doctoral programmes at the Faculty of Engineering of the University of Porto (FEUP).

Taking place on 30 June and 1 July, the event will offer a rich and diverse programme including keynote lectures, roundtable discussions, an exhibition and poster session, workshops, symposiums for all doctoral programmes, an awards ceremony, and various social events for participants. There will also be 34 company stands showcasing and promoting innovation and seeking to attract talent over the two days.

In this context, the Symposium in Informatics will be held on Tuesday morning, under the theme ‘Leading the Digital Transformation‘, and is organised by PhD students from the PhD Programmes in Informatics Engineering (PRODEI), Digital Media (PDMD) and Computer Science (MAP.i).

Following presentations of ongoing projects from the three programmes, the symposium will conclude with a lecture titled ‘Crowds and Graphics: Beyond Animation and Visual Effects‘, presented by Julien Pettré, a senior researcher at the Inria Centre at Rennes University.

One of the initiatives introduced in this year’s DCE edition is the ‘Venture Scientists‘ programme, which gives pre-selected doctoral students the chance to present an innovative research proposal in just five minutes. Presentations should focus on a product, process or service derived from the students’ research, highlighting its innovative potential and the corresponding implementation plan.

Seven proposals developed at FEUP will be in competition in this edition, to be evaluated by a panel of judges with extensive experience in academia and business: Wilson Caldeira (Visiting Professor at FEUP), Carrie Baptist (Chief Strategy Officer at Conception X), Raphael Stanzani (Head of Entrepreneurship Programmes at UPTEC), Bruno Azevedo (CEO of AddVolt) and Júlio Martins (CEO of Everythink).

We highlight the “BodyBoost” proposal by ProDEI student Ana Sofia Teixeira, who is currently in her second year of the PhD programme. The project involves developing a smart, discreet wearable device designed to be worn on the lower back. It continuously monitors posture and movements, silently identifying any bad posture habits. When necessary, it emits a slight vibration to stimulate quick and natural correction without interfering with the ongoing activity.

DCE25 has established itself as a platform of excellence for knowledge sharing, dialogue between different areas of knowledge and the promotion of synergies between research, innovation and society. It represents the scientific dynamics of FEUP and highlights the growing international profile of Portuguese engineering.

PhD Defence in Informatics Engineering: ”Towards Continuous Certification of Software Systems for Aerospace”

Candidate:
José Eduardo Ferreira Ribeiro

Date, Time and Location:
30th of June 2025, 14:30, Sala de Atos, Faculdade de Engenharia da Universidade do Porto

Title:
”Towards Continuous Certification of Software Systems for Aerospace”

President of the Jury:
Rui Filipe Lima Maranhão de Abreu (PhD), Full Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

Members:
Miguel Mira da Silva (PhD), Full Professor, Department of Computer Science and Engineering, Instituto Superior Técnico da Universidade de Lisboa;

João Miguel Lobo Fernandes (PhD), Full Professor, Departament of Informatics, Escola de Engenharia da Universidade do Minho;

João Carlos Pascoal Faria (PhD), Full Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;

João Gabriel Monteiro de Carvalho e Silva (PhD), Full Professor, Department of Informatics Engineering, Faculdade de Ciências e Tecnologia da Universidade de Coimbra (Co-Supervisor).

The thesis was supervised by Ademar Manuel Teixeira de Aguiar (PhD), Associate Professor of the Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

Abstract:

Since the publication of the Agile Manifesto in 2001, Agile methods have evolved to become the dominant approach in software development across diverse domains. However, their adoption in safety-critical systems development, such as aerospace, remains limited for reasons usually attributed to the stringent regulatory safety requirements imposed by domain-specific standards. This dissertation explores the applicability of Agile methods within the context of safety-critical aerospace software development, specifically under the guidelines of the DO-178C standard, and concludes that, contrary to common belief, Agile methods can be effectively used also in this context. The DO-178C standard, titled Software Considerations in Airborne Systems and Equipment Certification, is the principal certification guideline for aviation software for agencies such as Federal Aviation Administration (FAA) and European Union Aviation Safety Agency (EASA).

A key observation from discussions with professionals across different organizations and industries with strong safety requirements, including space, aerospace, railway, automotive, energy, and defence, is the widespread perception that traditional methods like the Waterfall model are indispensable, if not mandatory, for compliance and successful certification. This perception derives from the rigorous safety-related evidence required for certification. In aerospace software development, the minimal adoption of Agile methods and practices is attributed to the demands of DO-178C, regarded as a restrictive standard. However, contrary to this belief, DO-178C does not mandate any specific development method but instead provides guidelines and objectives to achieve the necessary safety-related evidence. This flexibility opens the possibility for Agile methods to be adapted to meet certification requirements while offering their well-documented advantages of incremental delivery and adaptability to changing requirements.

This research examines whether Agile methods, particularly the Scrum framework, can be effectively integrated into the development of safety-critical aerospace software systems while maintaining full compliance with DO-178C. The study introduces Scrum4DO178C, a novel Agile-friendly process tailored to address the specific challenges of aerospace software development, including its extensive verification and validation (V&V) efforts. Through a comprehensive review of literature, industry practices and data, as well as real-world insights from an industrial case study involving a critical aerospace project (Software Level A – Catastrophic), the research evaluates the feasibility and benefits of this approach. The case study demonstrates that Scrum4DO178C improves project performance, enhances responsiveness to changing requirements and reduces V&V efforts, in comparison with Waterfall, while fully complying with DO-178C.

The findings challenge the prevailing notion that Agile is inherently incompatible with safety-critical domains and suggest that when adapted thoughtfully, Agile methods can complement the rigorous standards requirements like DO-178C. By bridging the gap between Agile methods, practices and safety-critical development, this work advocates for a paradigm shift in developing safety-critical software, promoting a more adaptive, customer-centric approach. Specifically, this research highlights Agile’s capacity to accelerate knowledge acquisition through shorter delivery cycles and feedback loops, improve traceability, and manage late-stage requirement changes more efficiently, also in the aerospace domain.
Building on this foundational work, ongoing efforts are underway to enhance the Scrum4DO178C process through automation, enabling the automatic generation and reuse of outputs required for DO-178C compliance. Additionally, future research will extend these concepts to other aerospace standards and safety-critical domains, ensuring their applicability and compliance across diverse regulatory frameworks. Supported by collaborative initiatives with universities (e.g Master’s thesis projects at the Faculty of Engineering, University of Porto (FEUP) and the Informatics Engineering Departament of the University of Coimbra (UC)) and industry partners, this research aims to reshape industry perceptions of Agile’s role in safety-critical systems, fostering innovation and adaptability in these complex environments.

Keywords: Agile; Aerospace; DO-178C; FAA; Safety-critical; Software development.

PhD Defence in Informatics Engineering: ”An Optimization Strategy for Resource Allocation in Cyber Physical Production Systems”

Candidate:

Eliseu Moura Pereira

Date, Time and Location:

17th of June 2025, 10:00, Sala de Atos, Faculdade de Engenharia da Universidade do Porto

President of the Jury:

Carlos Miguel Ferraz Baquero-Moreno, Full Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto.

Members:

Pedro Nicolau Faria da Fonseca (PhD), Assistant Professor, Department of Electronics, Telecommunications and Computer Science, Universidade de Aveiro;

Paulo Jorge Pinto Leitão (PhD), Principal Coordinating Professor, Department of Electrical Engineering, Instituto Politécnico de Bragança;

André Monteiro de Oliveira Restivo (PhD), Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;

Gil Manuel Magalhães de Andrade Gonçalves (PhD), Associate Professor with Habilitation, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto (Supervisor).

The thesis was co-supervised by João Pedro Correia dos Reis (PhD), Assistant Researcher, Department of Electrical and Computer Engineering, Faculdade de Engenharia da Universidade do Porto.

Abstract:

Cyber-Physical Production Systems (CPPSs) integrate computation, communication, and control technologies, delivering the flexibility needed for dynamic shop floor reconfiguration and efficient manufacturing. A factory with higher shop floor flexibility has the advantage of higher product customization or changeover, when compared to traditional industries. They manifest this advantage mainly when the industry introduces a new product or because the shop floor produces highly variable products needing constant reconfiguration. Several manufacturers adopt such production philosophy, like in the automotive industry, where a high variability of car models and specifications requires different setups/configurations of the shop floor to manufacture them. In sequential production lines, like car assembly lines, reconfigurable CPPSs play an essential role because processing one product can affect the entire production performance, requiring a shop floor reconfiguration to optimize their execution. A significant challenge in CPPSs arises when reacting to changing conditions, such as new products or requirements, reconfiguration is needed. Current systems rely on manual intervention, leading to significant delays, especially in large industries where reprogramming hundreds of machines can take days or weeks. This thesis addresses this issue by proposing a platform that automatically optimizes software assignment to resources, speeding up development, deployment, and reconfiguration, enabling CPPSs to adapt to external disturbances quickly.

With the purpose of accelerating the development, reconfiguration, and execution of software in CPPSs, this thesis aims to optimize Function Blocks (FBs) assignment to the devices existent in an IEC 61499-based Cyber-Physical System (CPS), reducing the total execution time (reconfiguration time plus FB pipeline execution time). With this main goal, the thesis resulted in the development of 3 tools: 1) the Dynamic Intelligent Architecture for Software and Modular Reconfiguration (DINASORE), that enables the development, execution and manual reconfiguration of IEC 61499-based CPSs, 2) the Task Resources Estimator and Allocation Optimizer (TREAO),  that simulates and optimizes the tasks/FBs assignment to the CPS machines, recommending suitable software layouts for the CPS characteristics, and 3) the Task Assignment Optimization and Synchronization Engine (T-Sync), which integrates the previous two tools in a solution and optimizes in run-time the FBs assignment to the devices existent in an IEC 61499-based CPS.

Integrating these tools in T-Sync resulted in a differentiating solution because it 1) allows online FB assignment to optimize the CPS execution continuously and 2) improves the transparency and interoperability between FBs across IEC 61499-based devices. With this solution, the performance (total execution time) running FBs in reconfigurable CPSs improved by 30% in a simulated environment and 61% in a CPS. In addition to T-Sync improving total execution time, DINASORE enhances reconfiguration efficiency and flexibility, while TREAO streamlines CPS development by optimizing task and FB assignments to available resources. Besides the mentioned ones, during this thesis, other algorithms were implemented and tested for task assignment optimization, and other tools were developed to increase the interoperability and portability in CPSs. The future work envisions the automatic generation of FB pipelines from structured requirements, with formal specifications like UML diagrams, consequently integrating TREAO, manufacturing process simulators, and T-Sync to iteratively validate, optimize, simulate factory layouts, and deploy CPS software with enhanced flexibility and adaptability.

Keywords: Cyber-Physical Production Systems; IEC 61499; Machine Learning; Task Assignment.

PhD Defence in Informatics Engineering: ”Intelligent Ticket Management Assistant for Helpdesk Operations”

Candidate:

Leonardo da Silva Ferreira

Date, Time and Location:

13th of June 2025, 9:30, Sala de Atos, Faculdade de Engenharia da Universidade do Porto

President of the Jury:

Pedro Nuno Ferreira da Rosa da Cruz Diniz, PhD, Full Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto

Members:

Pedro Manuel Henriques da Cunha Abreu, PhD, Associate Professor with habilitation, Department of Informatics Engineering, Faculdade de Ciência e Tecnologia da Universidade de Coimbra;

Paulo Jorge Freitas de Oliveira Novais, PhD, Full Professor, Department of Computer Science, Escola de Engenharia da Universidade do Minho;

Carlos Manuel Milheiro de Oliveira Pinto Soares, PhD, Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;

Ana Paula Cunha da Rocha, PhD, Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto;

Daniel Augusto Gama de Castro Silva, PhD, Assistant Professor, Department of Informatics Engineering, Faculdade de Engenharia da Universidade do Porto (Supervisor).

The thesis was co-supervised by Professor Mikel Uriarte Itzazelaia, Associate Professor at the Escuela de Ingeniería de Bilbao, Universidad del País Vasco.

Abstract:

With the dynamic evolution of the internet, particularly in domains such as multimedia services, cloud computing, internet of things, virtualization, and artificial intelligence, companies have witnessed significant expansion in their market and services. However, this growth has also exposed numerous vulnerabilities that threaten the confidentiality, integrity, and availability of organizational and personal data. As information technology analysts work to address security system alerts, artificial intelligence has introduced new avenues for breaching security, ranging from simple, low-cost methods to highly sophisticated attacks. Low-cost approaches include phishing and password spraying, which exploit human error and weak password practices. In contrast, more complex threats include advanced persistent attacks and zero-day exploits, which require significant expertise and resources, often disrupting critical systems. Many organizations rely on cybersecurity helpdesk centers, internal or outsourced, to manage incidents. However, these centers often struggle to respond effectively due to data overload and a lack of qualified operators.

This dissertation addresses the shortage of skilled operators and the high volume of incidents in helpdesk operations by developing a ticket management assistant to support human operators in resolving incidents. The framework integrates a context-aware recommender system that identifies the fastest analyst-procedure pair for each incident and continually improves with each treatment followed. To ensure data privacy, this recommender system is trained using artificial data generated by a custom synthetic data generator. Furthermore, this thesis explores the possibility of enhancing this assistant with automated machine learning functionalities to predict incoming tickets. This feature could help managers anticipate workloads and proactively adjust the composition of the security teams.

The development of this framework is supported by the collaboration with a cybersecurity company, S21sec, which provides anonymized historical incident treatment data structures and taxonomies. However, synthetic data generation techniques are essential due to the absence of granular information on incident resolution and related parameters in the shared data set, which also requires privacy. The implemented generator builds artificial datasets that can mimic distributions similar to those observed in the real dataset while emulating real-world behaviours, including ticket prioritization, scheduling, and treatment.

The artificial data generator is evaluated for its efficiency in replicating real-world datasets using similarity measures such as Hellinger distance and Kullback-Leibler divergence. Furthermore, several ticket scheduling scenarios are explored, varying operators’ numbers and distribution across three work shifts. The results demonstrate that this framework can replicate ticket distributions and treatment durations observed in real datasets. Additionally, it allows for the simulation of real-world helpdesk operations, providing a solid foundation for exploring diverse operational contexts without compromising privacy. The analysis of the ticket scheduling consistently shows that scenarios characterized by a high shift imbalance and fewer operators lead to longer wait times and more tickets scheduled for later treatment.

The recommender system is assessed from two perspectives: scalability and impact on ticket treatment. The first phase uses various test datasets with different sizes and numbers of operators, analyzed with metrics such as the average recommendation time and memory usage. In contrast, the impact on ticket treatment is examined by considering improvements in ticket waiting times before being allocated to an operator and the response time required for their resolution, using different recommendation acceptance degrees. The results indicate that the number of operators the recommender system utilizes has a slightly larger impact on its scalability than the number of test tickets. Both features show a similar linear growth pattern regarding the referred metrics, but the number of operators has a larger slope. Integrating this recommender system into the ticket treatment reduced the average response time by 37.9\% to 45.1\% and the average wait time by 62.2\% to 63.2\%, assuming operators always accept the recommendations. With varying recommendation acceptance rates, the average wait time remains constant, while the response time improvement ranges from 0.4\% to 11.7\%.

The potential application of automated machine learning for predictive analysis is explored through a case study, comparing the system’s recommended team dimensionality decisions with expected outcomes. The case study evaluates the system based on prediction accuracy and its ability to suggest team size adjustments. Among the tested dataset distributions, models trained in three years of data outperformed those trained on four years, showing a better mean average error using real data on ticket frequency throughout the year. Regarding team dimensionality recommendations, including hiring or dismissing operators, the tool-based on automated machine learning frequently proposed decisions closely aligned with those that could have been proposed in the same period.

Collectively, these results show that the proposed framework can optimize ticket treatment workflows in real-world applications, leading to more efficient use of resources and reduced operational delays. Furthermore, its ability to simulate real-world operations without compromising privacy allows security operations centers to test several scenarios and refine their strategies.

Keywords: Helpdesk; Ticket; Cybersecurity; Synthetic Data; Recommendation Systems.

PhD Defence in Informatics Engineering: “Inmplode: A Framework to Interpret Multiple Related Rule-Based Models”

Candidate:

Pedro Rodrigo Caetano Strecht Ribeiro

Date, Time and Location:

13th of June 2025, 15:00, Sala de Atos, Faculdade de Engenharia, Universidade do Porto 

President of the Jury:

Rui Filipe Lima Maranhão de Abreu, PhD, Full Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto 

Members:

Johannes Fürnkranz, PhD, Full Professor, Department of Computer Science of the Institute for Application-Oriented Knowledge Processing at the Johannes Kepler University Linz, Austria;

José María Alonso Moral, PhD, Full Professor, Department of Electronics and Computing, Escuela Técnica Superior de Ingeniería de la Universidad de Santiago de Compostela, Spain;

José Luís Cabral de Moura Borges, PhD, Associate Professor, Department of Industrial Engineering and Management, Faculdade de Engenharia, Universidade do Porto;

João Pedro Carvalho Leal Mendes Moreira, PhD, Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto (Supervisor).

The thesis was co-supervised by Carlos Manuel Milheiro de Oliveira Pinto Soares, PhD, Associate Professor, Department of Informatics Engineering, Faculdade de Engenharia, Universidade do Porto. 

Abstract:

This thesis investigates the challenges and opportunities presented by the increasing trend of using multiple specialized models, referred to as operational models, to address complex data analysis problems. While such an approach can enhance predictive performance for specific sub-problems, it often leads to fragmented knowledge and difficulties understanding overarching organizational phenomena. This research focuses on synthesizing the knowledge embedded within a collection of decision tree models chosen for their inherent interpretability and suitability for knowledge extraction. For example, a company with chain stores or a university with diverse programs, each using dedicated prediction models (sales or dropout, respectively). While these localized models are important, a global perspective is valuable organization-wide. However, managing many operational models, especially for cross-program/store analysis, can be overwhelming.

A methodology framed within a comprehensive framework is introduced to merge sets of operational models into consensus models. These consensus models are directed towards higher level decision-makers, enhancing the interpretability of knowledge generated by the operational models. The framework, named Inmplode, addresses common challenges in model merging and presents a highly customizable process. This process features a generic workflow and adaptable components, detailing alternative approaches for each subproblem encountered in the merging process.

The framework was applied to four public datasets from diverse business areas and a case study in education using data from the University of Porto. Different model merging approaches were explored in each case, illustrating various process instantiations. The model merging process revealed that the resulting consensus models are frequently incomplete, meaning they cannot cover the entire decision space, which can undermine their intended purpose. To address the issue of incompleteness, two novel methodologies are explored: one relies on the generation of synthetic datasets followed by decision tree training. At the same time, the other uses a specialized algorithm designed to construct a decision tree directly from aggregated (i.e., symbolic) data.

The effectiveness of these methodologies in generating complete consensus models from incomplete rule sets is evaluated across the five datasets. Empirical results demonstrate the feasibility of overcoming the incompleteness issue, contributing to knowledge synthesis and decision tree modeling. However, tradeoffs were identified between completeness and interpretability, predictive performance, and the fidelity of consensus models.

Overall, this research addresses a critical gap in the literature by providing a comprehensive framework for synthesizing knowledge from multiple decision tree models, focusing on overcoming the challenge of incompleteness. The conclusions have implications for organizations seeking to use specialized models while maintaining a holistic understanding of the analyzed phenomenon.

Keywords: interpretability; rule-based models; model merging framework; decision trees; completeness.

DEI Talks | “How to hack the Turing trap with Trustworthy AI?” by José María Alonso Moral (CiTIUS-USC)

Tha talk “How to hack the Turing trap with Trustworthy AI?” will be presented June the 12th, at 18:00, in room B008, moderated by João Mendes Moreira (DEI).

Abstract:

In this talk, in addition to technical aspects (i.e., fundamentals and tools for developing and validating human-centric self-explaining technologies that are aimed at assisting in all phases of the design, analysis and evaluation of trustworthy intelligent systems), we will discuss Ethical, Legal, Socio-Economic and Cultural (ELSEC) implications of Artificial Intelligence. Special emphasis will be placed on how to certify if intelligent systems comply with European values.

About the Speaker:

José María Alonso Moral holds a M.S. and Ph.D. degrees in Telecommunication Engineering, both from the Technical University of Madrid (UPM), Spain, in 2003 and 2007, respectively. He is currently Associate Professor at the Department of Electronics and Computation of the CiTIUS-USC, Vice-Chair of the Task Force on “Explainable Fuzzy Systems” in the Fuzzy Systems Technical Committee of the IEEE Computational Intelligence Society (IEEE-CIS), Associate Editor of the IEEE Computational Intelligence Magazine (ISSN: 1556-603X) and the International Journal of Approximate Reasoning (ISSN:0888-613X), member of the IEEE-CIS Task Force on Fuzzy Systems Software, member of the IEEECIS SHIELD Technical Committee which is aimed at researching on Ethical, Legal, Social, Environmental and Human Dimensions of AI. He has published more than 190 papers in international journals, book chapters and conferences. His research interests include explainable and trustworthy artificial intelligence, computational intelligence, interpretable fuzzy systems, natural language generation, and the development of free software tools, etc.

The talk has free entrance, no need to register.

The last Lecture of Prof. A. Augusto de Sousa

Last Monday, 26th of May, the Auditorium José Carlos Marques dos Santos was filled to capacity for the final lecture by Prof. A. Augusto de Sousa, who was bidding farewell to a university career spanning more than 40 years.

Entitled ‘Computer Graphics and Multiple Realities: Explaining a Career‘, the lecture was attended by hundreds of students (both current and former), fellow teachers and staff, family and friends, who were present at this moment of emotional homage.

“Computer graphics allow us to represent and interact with virtual objects, as well as synthesise images of them. The tools, models and algorithms are countless and enable us to develop various realities, whether they are virtual, augmented or mixed. With some imagination, perhaps they can also model the reality of a university professor’s career.” – and so it was in this case. A university career that began in 1983 as an assistant trainee and went on to cover many different paths, always driven by a passion for Teaching and Pedagogy.

The session ended with a celebration featuring performances by TUNAFE, TEUP and Grupo de Fados da FEUP.

The final lecture was broadcasted live and is available to watch here.