Healthy and sustainable development of sports economy based on artificial intelligence and mental model

Yue Liu, Bo Dong, Xiangcheng ZengFrontiers in Psychology 2022 · DOI

Abstract

In the modern era, the sports economy has transitioned from a niche sector to a primary driver of global economic growth. However, ensuring the healthy and sustainable development of this industry requires balancing rapid expansion with social and environmental factors. This paper introduces a predictive framework that utilizes Deep Learning (Restricted Boltzmann Machines) and Support Vector Machines (SVM) to construct a "mental model" of the sports economy. By analyzing indicators from a Chinese province between 2017 and 2021, the study reveals that an AI-optimized sustainable development path can enhance growth indicators by approximately 20% over traditional models.

Introduction

The sports economy is a complex economic form characterized by specialized organizations, consumer participation, and profit-oriented operations. As the connection between sports and the broader economy deepens, the challenge shifts from achieving growth to maintaining it sustainably. Traditional economic models often struggle with the non-linear complexities of this sector. This research addresses the gap by integrating artificial intelligence with the psychological concept of "mental models" to create a more holistic predictive tool.

Methodology

The researchers employ a hybrid AI approach to bridge the gap between industry organization (the Implementation Model) and public perception (the Represented Model).

The AI Pipeline

The technical framework consists of two primary components:

  1. Restricted Boltzmann Machines (RBM): This deep learning algorithm is used for feature extraction. RBMs are effective at identifying hidden structures within complex datasets without requiring explicit feedback loops, making them ideal for mapping the non-linear relationships in economic data.
  2. Support Vector Machines (SVM): Following feature extraction, SVMs are utilized to classify and predict development states. The algorithm identifies the "optimal separating hyperplane" to distinguish between standard growth trajectories and optimized, sustainable paths.
The dual-mode identification and study framework for pattern classification in the sports economy model.

Mental Model Integration

The "mental model" serves as the conceptual bridge. It ensures that the "Performance Model"—the actual outcome of economic policies—aligns with both the developer's intent and the user's experience. By training the AI on these intersecting perspectives, the model can forecast outcomes that prioritize long-term public engagement alongside financial gain.

Data and Indicators

The study focuses on "Province A" in China, analyzing a time-series dataset from 2017 to 2021. The model tracks seven key indicators:

  • Economic: Regional GDP and total output of the sports industry.
  • Social/Infrastructure: Number of sports practitioners, local financial expenditures on sports, the volume of development policies, public participation rates in physical exercise, and the availability of fitness venues.

During the study period, the province's sports market grew from 55.3 billion yuan in 2017 to 103.5 billion yuan in 2021, though the rate of growth showed signs of natural deceleration.

Key Insights: The Three Pillars of Sustainability

The research defines sustainability through three core laws:

  • Equity: Ensuring horizontal fairness for the current generation and vertical fairness for future generations.
  • Continuity: Maintaining stable and orderly development over time.
  • Coordination: Achieving compatibility between economic goals, natural resources, and the environment.

A critical finding is the strong correlation between public participation and GDP. The model suggests that public health and exercise frequency are not just social benefits but direct drivers of economic value.

Results and Discussion

The implementation results demonstrate a significant "Optimization Rule." When the AI model forecasted future trends, it found that under "normal" conditions, growth rates for total output and financial expenditure would likely stagnate.

However, under the "healthy and sustainable" framework, the growth rates of various indicators increased significantly. The sustainable model maintained a stable growth rate of approximately 10%, whereas the normal development scenario predicted a sharper decline as the market matured. Overall, the sustainable path offered a 20% optimization across the tracked variables.

Conclusion and Future Work

While the study provides a robust proof-of-concept for AI-driven economic planning, it is limited by its geographic focus on a single province and a relatively short five-year data window. Future research will aim to incorporate multi-regional comparisons and longer-term datasets to refine the mental model's accuracy for global applications. By leveraging AI to balance economic and social indicators, policymakers can ensure that the sports industry remains a healthy, sustainable pillar of the global economy.

Summary updated

Abstract

In recent years, sports have achieved rapid development worldwide, and the global economy has been significantly improved and improved. With the in-depth development of the two, the connection between sports and the economy has also become closer. Sports economy is a new type of economic form bred by specialization of sports organization, participation in consumerization, and profit-oriented operation under the condition of market economy. And the development of sports economy cannot be developed at once; it needs healthy and sustainable development. In order to find a better way to study the healthy and sustainable development of sports economy, this paper uses deep learning network algorithm and supports vector machine learning algorithm to build a mental model. It then uses the model to analyze various indicators of the sports industry in a province in China. This article is looking for information and summarizes the province's sports data from 2017 to 2021. The sports indicators of this experiment include regional GDP, total output of sports industry, sports practitioners, local financial sports expenditures, the number of policies, the number of people participating in physical exercise, and fitness venues and facilities. The realization results show that these variables develop at a relatively small rate under normal conditions, and then predict the data in the next few years under the healthy and sustainable development of the next few years through the mental model. The growth rates of various indicators of the sports economy have increased significantly, and they have been optimized by about 20% compared with the normal development.
BibTeX
@inproceedings{liu2022healthy,
  author = {Yue Liu and Bo Dong and Xiangcheng Zeng},
  title = {{Healthy and sustainable development of sports economy based on artificial intelligence and mental model}},
  booktitle = {Frontiers in Psychology},
  year = {2022},
  doi = {10.3389/fpsyg.2022.956682},
}
Figure