INTACT - Détection des Intentions, Prédiction de l’Action

DEMO : please contact me, at alda DOT mari “@” cnrs DOT fr | alda mari

PI INTACT

• Human Sciences. Linguistics : Alda Mari (CNRS/ENS/EFPS), Language, Thought and Behavior Team (Project Lead)
• Computer Science. Farah Benamara (IRIT and University of Toulouse), Mélodi Team

Distinctions

• Alda Mari : Innovation Medal CNRS 2025
• Alda Mari : Innovation Price in Human Sciences 2024

Members
Permanent

• Colonel Laurent Leygue (DGSCGC, Head of the Strategy and Foresight Mission)
• Gaëtan Chevalier (Engineer, PhD candidate, UTT)
• Véronique Moriceau (IRIT and University of Toulouse), Mélodi team
• Valentin Taffin TTT Toulouse
• Jérôme Lelasseu TTT Toulouse

Non-permanent

• 4 engineers, 5 PhD students, +15 Master Students

Project Phases and Funding Institutions

• INTACT I : CHEMI 2018–2019
• INTACT II : FIESPI 2020–2022
• NTACT III : CNRS Maturation Program, CNRS Innovation Laureate Alda Mari
• NTACT IV Maturation with SATT, TTT

What do we do know ?
We have a patent and a running software ! Please, contact me at my CNRS address

Why INTACT is unique ?
Because it resorts to a very sophisticated semantic based annotation that considers speaker’s evaluation and perspective, based on work of Giannakidou and Mari (2021,2024) in semantics and pragmatics. Emergency is thus treated not only as domain dependent but as a stance, anchored to individuals.

What we have achived ?

Use social media effectively during active crises

• Recognize that social networks have become a key communication channel during emergencies.
• Develop methods to exploit information shared on social media while a crisis is unfolding.

Improve crisis communication

• Help authorities communicate more efficiently with the public through social media.
• Address the challenges and opportunities created by digital communication platforms.

Increase societal resilience

• Enhance society’s ability to respond to and recover from crisis-triggering events.
• Leverage information circulating on social networks to support resilience.

Support crisis management

• Provide tools, methods, or analyses that help decision-makers manage crises more effectively.
• Improve situational awareness by monitoring and understanding online communications.

Benefit civil security and public order

• Increase the effectiveness of crisis-management work for emergency services, civil protection agencies, and public-security organizations.

Selected publications

Giannakidou, A. and Mari, A. (2021). Truth and Veridicality in Grammar and Thought. The University of Chicago Press.

Giannakidou, A. and Mari, A. (2026). Modal Sentences. Cambridge University Press.

Kozlowski, D., Lannelongue, E., Saudemont, F., Benamara, F., Mari, A., Moriceau, V., & Boumadane, A. (2020). A three-level classification of French tweets in ecological crises. Information Processing & Management, 57(5), 102284.

Laurenti, E. Nils, B., Benamara, F., Alda, M., Moriceau, V., & Camille, C. (2022). Speech acts and communicative intentions for urgency detection. In Proceedings of the 11th Joint Conference on Lexical and Computational Semantics. (pp. 289-298).