Neuro-symbolic approach for querying BIM models
Approche neuro-symbolique pour requêter les modèles BIM
Résumé
Integrating technologies like the Internet of Things (IoT) and artificial intelligence (AI) has transformed the smart building domain, enabling innovative services. The data gathered throughout a building's lifecycle, including construction and usage, create a comprehensive database for efficient management and improvement, mainly through Building Information Modelling (BIM). However, accessing and utilising these data can be challenging for users, as it often requires specialised query skills. Improving information accessibility involves intuitive and user-friendly query interfaces. Traditional approaches struggle to provide the flexibility and adaptability needed to keep up with evolving building information and user language expressions. To address these challenges, modern and scalable methods from symbolic and numeric artificial intelligence offer a solution. These systems can handle varying queries, adapt to changing data, and provide more user-friendly interactions, aligning better with user expectations in the industry.