Article: Dynamic scheduling and recommendation of cultural tourism resources using AlphaEvolve and graph convolutional networks Journal: International Journal of Information and Communication Technology (IJICT) 2026 Vol.27 No.82 pp.35 - 55 Abstract: Static scheduling inadequately addresses dynamic tourist preferences and spatiotemporal resource heterogeneity in cultural tourism. This study proposes a hybrid framework integrating AlphaEvolve with graph convolutional networks (GCNs). The model extracts topological features via GCNs and constructs dynamic strategy spaces through AlphaEvolve, learning resource-demand matching from real-time behavioural and environmental data. Results demonstrate 82.7% overall recommendation accuracy (89.6% historical, 82.4% real-time, 76.1% dynamic event scenarios). Response time decreases from 12.5 to 4.8 minutes, with 91.7% peak-hour resource utilisation and 3.1% wait timeout rate. User satisfaction improves substantially: content relevance (71.2 → 89.5), timeliness (63.8 → 91.2), short-term conversion (12.3% → 28.7%), long-term conversion (8.9% → 22.4%), and daily usage reaches 32.1 minutes. Strategy evaluation error converges below 0.1, enhancing scheduling stability by 33%. Inderscience Publishers - linking academia, business and industry through research

Title: Dynamic scheduling and recommendation of cultural tourism resources using AlphaEvolve and graph convolutional networks

Authors: Meng Wang

Addresses: School of Cultural Tourism, Henan Institute of Economics and Trade, Zhengzhou, 450000, China

Abstract: Static scheduling inadequately addresses dynamic tourist preferences and spatiotemporal resource heterogeneity in cultural tourism. This study proposes a hybrid framework integrating AlphaEvolve with graph convolutional networks (GCNs). The model extracts topological features via GCNs and constructs dynamic strategy spaces through AlphaEvolve, learning resource-demand matching from real-time behavioural and environmental data. Results demonstrate 82.7% overall recommendation accuracy (89.6% historical, 82.4% real-time, 76.1% dynamic event scenarios). Response time decreases from 12.5 to 4.8 minutes, with 91.7% peak-hour resource utilisation and 3.1% wait timeout rate. User satisfaction improves substantially: content relevance (71.2 → 89.5), timeliness (63.8 → 91.2), short-term conversion (12.3% → 28.7%), long-term conversion (8.9% → 22.4%), and daily usage reaches 32.1 minutes. Strategy evaluation error converges below 0.1, enhancing scheduling stability by 33%.

Keywords: cultural and tourism resource scheduling; AlphaEvolve algorithm; graph convolutional network; GCN; dynamic recommendation; multi-objective optimisation.

DOI: 10.1504/IJICT.2026.155116

International Journal of Information and Communication Technology, 2026 Vol.27 No.82, pp.35 - 55

Received: 26 Feb 2026
Accepted: 16 Apr 2026
Published online: 27 Jul 2026 *

Title: Dynamic scheduling and recommendation of cultural tourism resources using AlphaEvolve and graph convolutional networks

Keep up-to-date