Machine Learning Reveals How Campus Green Spaces Impact Student Well-being in Tehran

Revolutionizing Campus Design Through AI-Driven Aesthetic Analysis

In a groundbreaking study from Tehran’s academic institutions, researchers are leveraging machine learning to decode the complex relationship between campus environments and student aesthetic preferences. This innovative approach combines environmental psychology with advanced computational models to identify which landscape elements most significantly contribute to students’ psychological restoration and visual satisfaction.

The research team developed a comprehensive methodology that bridges traditional survey techniques with cutting-edge artificial intelligence. By analyzing 100 carefully curated images from four major Tehran universities—Amirkabir University of Technology, University of Tehran, Shahid Rajaee Teacher Training University, and Tarbiat Modares University—the study establishes a new framework for understanding how students interact with and benefit from their campus surroundings.

Methodological Innovation: From Field Surveys to Machine Learning

The research employed a sophisticated mixed-methods approach beginning with spatial analysis and on-site surveys to identify student-preferred rest spots. Researchers achieved data saturation after surveying 443 students across the four campuses, ensuring comprehensive coverage of preferred locations. The photographic documentation followed rigorous protocols, with images captured under consistent lighting conditions and from standardized perspectives to eliminate variables that could skew aesthetic judgments.

What sets this study apart is its integration of environmental psychology principles with machine learning applications. The research team grounded their variable selection in established theories including Attention Restoration Theory (ART) and Stress Reduction Theory (SRT), then used these variables to train multiple machine learning models for predicting aesthetic preferences.

Environmental Variables: The Building Blocks of Aesthetic Preference

The study extracted and analyzed, comprehensive coverage, eighteen environmental variables previously associated with psychological restoration, spanning three main categories:

• Architectural features: Buildings, pathways, sculptures, and overhead shading systems
• Landscape elements: Tree cover, soft landscapes, waterscapes, flowers, natural stone, lawns, vertical greenery, and shrubs
• Campus facilities: Environmental amenities and seating facilities

Each variable was meticulously measured using both quantitative and qualitative methods. Area coverage calculations were performed using AutoCAD and Excel, while more subjective elements like landscape complexity were assessed by expert panels using specially developed evaluation scales., according to market developments

Machine Learning Models: Predicting Aesthetic Responses

The research team implemented a diverse set of machine learning algorithms to analyze the relationship between environmental features and aesthetic preferences:

• Support Vector Regression (SVR) for handling nonlinear relationships
• Random Forest (RF) for robust feature importance analysis
• Multilayer Perceptron (MLP) for capturing complex patterns
• Ensemble Learning Models for improved prediction accuracy

The models were trained on 80% of the dataset and tested on the remaining 20%, with careful attention to preventing data leakage through proper standardization techniques. This multi-model approach allowed researchers to compare performance and identify the most reliable predictors of aesthetic preference.

Key Findings: What Makes Campus Spaces Restorative?

Preliminary analysis reveals several critical factors that significantly influence students’ aesthetic preferences and psychological restoration:

Natural elements emerge as paramount, with tree cover, water features, and plant diversity consistently ranking high in predictive importance. The presence of seating facilities and environmental amenities also proved crucial, highlighting the importance of both comfort and functionality in campus design.

Interestingly, the study found that moderate landscape complexity—neither too simple nor overwhelmingly intricate—tended to generate the most positive aesthetic responses. This finding aligns with previous research suggesting that environments offering moderate visual stimulation are most conducive to attention restoration.

Implications for Campus Planning and Urban Design

This research represents a significant advancement in evidence-based campus design. By identifying specific environmental features that contribute to student well-being, university administrators and landscape architects can make data-driven decisions about campus development and renovation.

The machine learning approach offers a scalable framework that could be adapted to other educational institutions worldwide, potentially revolutionizing how we design learning environments that support both academic achievement and mental health.

As universities increasingly recognize their role in supporting student wellness, studies like this provide crucial empirical evidence for creating campuses that aren’t just functional, but genuinely restorative. The integration of environmental psychology with machine learning represents an exciting frontier in campus design—one that promises to create educational spaces that actively contribute to student success and well-being.

This article aggregates information from publicly available sources. All trademarks and copyrights belong to their respective owners.

Note: Featured image is for illustrative purposes only and does not represent any specific product, service, or entity mentioned in this article.