Keynotes
Bio sketch
Born in 1947, Robert Laurini (aka Roberto) holds two doctorates, one in 1973, and the other in 1980, both in information technologies awarded by the Claude Bernard University of Lyon, France. He speaks fluently French, English, Italian and Spanish.
Throughout his career, he primarily worked at INSA-Lyon (University of Lyon), eventually achieving the status of distinguished professor. However, in 1976-77, he spent an entire year as a research associate at the Martin Centre of the University of Cambridge in the United Kingdom. In addition, in 1986-87, he served as a visiting professor at the University of Maryland, College Park, USA. Between 1995 and 2005, he held a part-time position at IUAV University in Venice, Italy. Since 2011, he has retired and holds the title of professor emeritus.
During his career, he worked overall on computing aspects of geographic information systems, namely spatial data indexing, geometric quality control, geographic database consistency, updating through aerial photos, visual summaries, chorem generation from spatial data mining, geographic ontologies, sensor data organization and indexing, etc. In total, he supervised or co-supervised 44 PhD students in those domains. His more recent works have been on geographic knowledge, especially targeted to smart cities and territorial intelligence.
He wrote several books regarding geographic information systems, especially for urban and environmental planning and the last one is “Geographic Knowledge Infrastructure: Applications to Territorial Intelligence and Smart Cities” published in 2017. He authored or co-authored more than 250 papers in various journals and conferences.
He was invited to give seminars in many countries, especially in Italy, Morocco, Algeria, Argentina, and Mexico. He was a member of PhD committees more than 100 times in 19 countries.
During the nineties, he was president of the ACMGIS steering committee (now SIGSpatial), vice-president of UDMS, European editor of Computers, Environment and Urban Systems, and associate editor of the Journal of Visual Languages and Computing.
In 2009, he founded the NGO named “Universitaires Sans Frontières/Academics Without Borders” the scope of which is to help modernize universities in developing countries. In this NGO, he is now in charge of several universities in Latin America, especially for the training of PhD students and young researchers for scientific writing, and for lab heads in the definition and validation of promising novel research lines.
For more details see www.laurini.net/robert.
Abstract
Knowledge Modeling for Urban Planning
Decisions regarding the future of a city are based on 4 types of knowledge bundles, awareness of the local context, current regulations about urban planning, planners’ objectives and information coming from outside experiences. In order to make reasoning with those knowledge bundles, a language has been created based on rules and set theory. Those rules combine urban objects (buildings, streets, parcel, etc. together with their ontology), relationships between them (topologic, cardinal, etc.), mathematical models (static or dynamic) and gazetteer (dictionary of placenames). Sometimes knowledge chunks are not based on previous experiences, but derived from anticipations, known as feedforward knowledge. The scope of this key address will be, after having rapidly presented some applications, (i) to describe the grammar of this language, (ii) to give several examples and (iii) to explain how to use them in decision making. Finally, some remarks will be given to define territorial intelligence by mixing adequately artificial intelligence and collective human intelligence.
Dr. Robert Laurini
Professor Emeritus in Information Technologies
Franklin Tchakounte
Associate Professor at the University of Ngaoundere in Cameroon
Bio sketch
Franklin Tchakounte holds a PhD in Computer Science from the University of
Bremen, Germany, and serves as an Associate Professor at the University of
Ngaoundere in Cameroon and affiliated to Kesmonds International University,
USA. With over two decades of experience in digital transformation,
cybersecurity, and data intelligence, he founded and chairs the Cybersecurity
with Computational and Artificial Intelligence Group. His academic journey
includes over 40 publications—ranging from research papers to conference
proceedings and book chapters—and certifications in areas such as Cybersecurity
Management, Artificial Intelligence Strategies and Governance.
He is actively engaged in national and international professional societies and
have been honored with awards, fellowships, and research grants in recognition
of his contributions. He co-invented AI-based platforms like eyeandears.com,
which leverages AI to address misinformation, and the Cybersecurity Academy,
designed to promote awareness about digital safety, relying on Generative AI, for
all age groups.
Read more..
Dedicated to youth empowerment and resilience, I recently launched the
ResilientMinds initiative, which fosters leadership and personal growth among
youth in underprivileged communities. He is also a pioneer of responsible AI in
Cameroon. I Founded the International Conference on Safe, Secure, Ethical, and
Responsible Technologies and Emerging Applications (SAFER-TEA), and
represents Cameroon in the Responsible AI Network–Africa. As a digital
transformation expert for the Regional Conference of Rectors of Central African
and Great Lakes Universities, he advises on the responsible integration of
generative AI in education. He is recipient of the 2025 AI ChangeMaker Award by
the IA Cameroon association.
To support youth development, he launched the ResilientMinds initiative, which
fosters leadership and personal growth among young people in underserved
communities. He has also represented Cameroon in regional and global
discussions on emergent technologies and digital economies. He co-edited the
book Global Perspectives on the Applications of Computer Vision in Cybersecurity
and contributed to the elaboration of ICESCO Charter on AI Ethics. He is the
Principal Investigator of a three-years U.S.-funded project by DEVCOM ARL and
ONR Global, for skilling of the next generation of African leaders in cybersecurity
and AI.
Abstract
Mobile money (MM) is a critical financial service for innovations, particularly in
underserved areas in Africa, yet its rapid adoption has made it a prime target for
malicious actors exploiting social engineering techniques such as phishing. While
existing countermeasures address technical vulnerabilities, they often overlook
the dynamic and sequential nature of attacker–victim interactions. In this talk, we
will present a multi-year’s study in my research group, that proposes a novel
learning-based framework for characterizing mobile phishing attacks, leveraging
both Reinforcement Learning (RL) and Imitation Learning (IL). Initially, RL
algorithms such as Q-learning and Deep RL (DQN, A2C, DDPG) will be explored,
to modeling attacker behavior through interaction with a Markov Decision
Process. Although these approaches offer fast convergence, they require
carefully defined reward functions, which can be difficult to design in complex
social environments. To overcome this, we implemented two IL methods:
Behavior Cloning and Dataset Aggregation (DAgger), trained on real attack
sequences from Orange and MTN mobile money services in Cameroon. The
conducted experiments reveal that DAgger significantly outperforms Behavior
Cloning, generating accurate attacker actions after only two iterations.
Compared to Q-learning, IL offers greater interpretability and behavioral
consistency, particularly when expert demonstrations are available. This project
lays the foundation for developing adaptive and intelligent defenses against
phishing, and opens pathways for future extensions including multi-agent
simulations and causal reasoning, and it opens way forward for research
collaboration.
