Designing a Neural‑Network Model for Sport Entrepreneurship in the Tourism Context

Articles in Press

Document Type : Original Article

Authors

1 Student of Islamic Azad University \Qazvin \ Iran

2 Department of Sports Management, Faculty of Management and Accounting, Islamic Azad University, Qazvin Branch, Iran

10.22034/sms.2025.143670.1501

Abstract

Aiming to deepen understanding of the factors that shape sport entrepreneurship within Iranian tourism, this study develops a neural‑network model to explain and predict the phenomenon. The applied research employed an exploratory sequential mixed‑methods design. In the qualitative phase, data were gathered through interviews with 12 academic and industry experts and a literature review, and then analyzed thematically. In the quantitative phase, 282 sport‑ and tourism‑sector practitioners completed a researcher‑made questionnaire containing 52 items across nine constructs; content validity was verified via the Lawshe method and reliability via Cronbach’s alpha (0.86). Data were examined using exploratory factor analysis (EFA) and a multilayer perceptron (MLP) neural network. Before entering the network, all variables were normalized to the [0–1] range. The analysis identified nine principal constructs—cultural factors, economic factors, consumer/user, online advertising, satisfaction, credibility, trust, customer‑relationship management, and sport entrepreneurship in tourism. The nine‑factor structure explained 61.83 % of the total variance. The developed MLP model displayed high predictive power and confirmed a direct, positive, and significant impact of all eight independent constructs on sport entrepreneurship. Cultural (normalized importance = 100%) and economic factors (99.4%) were the most influential. The model offers a conceptual and practical framework for analyzing and fostering sport entrepreneurship. The chief innovation lies in designing and validating the first comprehensive predictive model for this domain.

Keywords

Main Subjects

Sports Marketing Studies

Articles in Press, Accepted Manuscript
Available Online from 24 July 2025

Files

Share

How to cite

Statistics