Tipus de document

Article

Versió

Versió acceptada

Data de publicació

Tots els drets reservats

Si us plau utilitzeu sempre aquest identificador per citar o enllaçar aquest document: https://hdl.handle.net/2445/232735

Fractal Deep Neural Networks for Forecasting the Preferences on Sustainable Academic Conferences

Títol de la revista

Director/Tutor

Contribució addicional

ISSN de la revista

Títol del volum

Resum

This research aimed to construct a predictive model to estimate attendance preferences for sustainable academic conferences. The model employed fractal-based neural network architectures within a machine learning framework, specifically utilizing Fractal Convolutional Neural Networks (Fractal CNN), Fractal Deep Recurrent Neural Networks (Fractal DRNN), and Fractal Long Short-Term Memory (Fractal LSTM) networks. Although conventional neural networks have been widely used across multiple disciplines, this study introduced an innovative application by embedding fractal dynamics into the network structure to assess individual attitudes toward sustainability in academic event settings. Additionally, the research examined how demographic attributes and broader environmental awareness within participants’ countries influence event-related preferences. Data collected from 532 respondents representing 68 nations enabled the study to investigate connections between national environmental engagement and individual behavioral responses. The analysis found that age and gender were among the most influential demographic predictors of support for sustainable conferences. Women and individuals aged 30 and above expressed stronger inclinations toward environmentally responsible practices. Other findings indicated that participants consistently held positive views on sustainable initiatives. Key aspects such as venue choice, food services, use of eco-friendly materials, and lodging arrangements were identified as central to shaping these attitudes, whereas transport-related elements appeared less influential. Among the tested models, the Fractal CNN achieved the highest predictive accuracy and demonstrated superior performance in minimizing forecasting errors. These results emphasize the strength of fractal neural approaches in capturing complex behavioral patterns and highlight their relevance for advancing predictive tools in sustainability-focused event planning.

Citació

Citació

SANTOS, José António C., ALAMINOS AGUILERA, David and FERNANDEZ-MIGUELEZ, Sergio M. Fractal Deep Neural Networks for Forecasting the Preferences on Sustainable Academic Conferences. Fractals-Complex Geometry Patterns and Scaling in Nature and Society. 2026. ISSN 0218-348X. [consulted: 18 of November of 2026]. Available at: https://hdl.handle.net/2445/232735

Exportar metadades

JSON - METS

Compartir registre