{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,16]],"date-time":"2026-11-16T16:02:29Z","timestamp":1794844949939,"version":"4.9.1"},"reference-count":0,"publisher":"IOS Press","isbn-type":[{"value":"9781643686660","type":"electronic"}],"license":[{"start":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T00:00:00Z","timestamp":1780358400000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,6,2]]},"abstract":"<jats:p>Traffic forecasting requires modeling complex spatiotemporal dependencies that evolve across multiple temporal scales and heterogeneous spatial relationships. In this paper, we present the architectural framework and mathematical formulation of MS-BGFN, a novel architecture that integrates multi-scale temporal attention with dynamic multi-graph convolutions. The model processes traffic signals at four complementary temporal resolutions (5-minute, hourly, daily, and weekly). By mathematically decoupling traffic dynamics into resolution-specific spectral filters, we address the temporal blur inherent in single-scale models, effectively capturing both short-term dynamics and long-term periodic patterns. To model spatial dependencies, MS-BGFN learns both topological and behavioral adjacency matrices, enabling adaptive graph representations that reflect evolving traffic interactions. A multi-graph fusion module combines these predefined and data-driven adjacency matrices through learnable graph-level gating. Finally, we present a comprehensive complexity analysis demonstrating that MS-BGFN maintains O(K|E|) efficiency, making it highly suitable for real-time, city-scale deployment. This architecture provides a scalable framework for multi-scale traffic forecasting in intelligent transportation systems.<\/jats:p>","DOI":"10.3233\/aise260019","type":"book-chapter","created":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T08:19:48Z","timestamp":1781252388000},"source":"Crossref","is-referenced-by-count":0,"title":["Multi-Scale Behavioral Graph Fusion Network for Traffic Forecasting"],"prefix":"10.3233","author":[{"given":"Doha","family":"Bakir","sequence":"first","affiliation":[{"name":"Laboratory of Mathematics and Informatics, Faculty of Sciences, Hassan II University of Casablanca, Casablanca, Morocco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Khalid","family":"Moussaid","sequence":"additional","affiliation":[{"name":"Laboratory of Mathematics and Informatics, Faculty of Sciences, Hassan II University of Casablanca, Casablanca, Morocco"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zouhair","family":"Chiba","sequence":"additional","affiliation":[{"name":"Laboratory of Mathematics and Informatics, Faculty of Sciences, Hassan II University of Casablanca, Casablanca, Morocco"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"7437","container-title":["Ambient Intelligence and Smart Environments","Intelligent Environments 2026: Combined Workshop Proceedings"],"original-title":[],"link":[{"URL":"https:\/\/ebooks.iospress.nl\/pdf\/doi\/10.3233\/AISE260019","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T08:19:48Z","timestamp":1781252388000},"score":1,"resource":{"primary":{"URL":"https:\/\/ebooks.iospress.nl\/doi\/10.3233\/AISE260019"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,2]]},"ISBN":["9781643686660"],"references-count":0,"URL":"https:\/\/doi.org\/10.3233\/aise260019","relation":{},"ISSN":["1875-4163","1875-4171"],"issn-type":[{"value":"1875-4163","type":"print"},{"value":"1875-4171","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,2]]}}}