Thematic Session 23

THEMATIC SESSION #23

Malvina Halilaj

University of Tirana

• malvina.halilaj@fshn.edu.al

Paolo Sernani

University of Macerata, Italy

• paolo.sernani@unimc.it

Paolo Contardo

Polytechnic University of Marche, Italy

• p.contardo@univpm.it

Erisa Bekteshi

University of Tirana

• erisa.bekteshi@fshn.edu.al

Violence is an increasingly critical social issue that severely impacts the well-being of its victims, particularly in cases of domestic abuse involving women, minors, the elderly, and marginalized individuals. Many violent incidents, including physical assaults and verbal aggression, remain underreported, making recognition and intervention challenging. Physical violence can manifest in various forms, from minor physical altercations to severe assaults, while non-physical violence, such as verbal abuse on social media, also poses significant risks.

This thematic session aims to explore AI-driven solutions for detecting and measuring violence. It invites research on the use of video and audio analysis for recognizing physical aggression, as well as lexical tools for identifying offensive and harmful content on digital platforms. Topics of interest include gender-based violence, child abuse, and the detection of explicit content such as child pornography and revenge porn. By leveraging artificial intelligence, this session seeks to advance methodologies for early detection and intervention, contributing to a safer and more secure society.

Extended version of accepted papers will be considered for publication in a special issue of "Intelligenza Artificiale".

Topics of interest for this Special Session include but are not limited to:

• Gender-based violence
• Violence against minors
• Recognition of violence from video and/or audio
• Recognition of texts with violent content
• Recognition of nudity (child pornography, revenge porn)
• AI for crime prevention
• Hate speech detection
• Social media content analysis
• Cyberbullying detection
• Deep learning for violence recognition
• Surveillance system enhancement with AI
• Forensic AI applications for violence detection
• Ethical AI in crime detection

Malvina Halilaj is an Assistant Professor at the University of Tirana, Department of Computer Science, and is currently pursuing a PhD at Università Politecnica delle Marche (UNIVPM). She holds a Master of Science in Computer Science. Her research interests include artificial intelligence, specifically its applications in countering crime , she is involved in research on violence detection in videos and AI-driven social network analysis. Her work aligns with projects under the collaboration agreement between AIRTLab (Artificial Intelligence & Real-Time Laboratory) and key institutional partners, focusing on the application of artificial intelligence .

Paolo Sernani is currently Assistant Professor at the Department of Law of the University of Macerata. He received a Ph.D. degree in Information Engineering at Università Politecnica delle Marche in March 2016, defending a thesis entitled “Design and virtualization of intelligent systems for the management of assistive environments” dealing with the application of multi-agent systems to manage assistive smart environments. His main research interests include expert systems, deep learning, and hybrid techniques.

Paolo Contardo has completed his Ph.D. at Università Politecnica delle Marche ,with a dissertation titled "Machine Learning in Forensics: Applications of Convolutional Neural Networks in Fingerprint Scanning, Violence Recognition, and Signal Detection." During his Ph.D., he worked on the projects "Dactyloscopy 2.0," "Fotosegnalamento 2.0," and "Violence Detection in Videos," which were developed within the framework of the collaboration agreement between Università Politecnica delle Marche (UNIVPM) CARMELO Inter-Departmental Research Center (Center for Advanced Research on Measurements for Engineering and Life Optimization) and the "Ministero dell’Interno, Dipartimento della Pubblica Sicurezza, Direzione Centrale Anticrimine della Polizia di Stato" . His research focused on applying artificial intelligence and deep learning to forensic investigations, enhancing security and crime prevention through innovative AI-driven solutions.

Erisa Bekteshi is an Assistant Professor at the University of Tirana, where she teaches informatics-related subjects. She is also a software developer with a strong background in informatics and a PhD student in Complex Systems Engineering at the Department of Innovation Engineering, focusing on AI-enabled indoor navigation systems. Her research explores the application of artificial intelligence in navigation and smart environments. Additionally, she is engaged in projects related to AI-driven violence detection, leveraging machine learning and data analysis to enhance safety in digital and physical spaces.