Abstract:Due to increased reliance on private vehicles and growing travel demand, parking remains a longstanding urban challenge globally. Quantifying parking perceptions is paramount as it enables decision-makers to identify problematic areas and make informed decisions on parking management. This study introduces a cost-effective and widely accessible data source, crowdsourced online reviews, to investigate public perceptions of parking across the U.S. Specifically, we examine 4,987,483 parking-related reviews for 1,129,460 points of interest (POIs) across 911 core-based statistical areas (CBSAs) sourced from Google Maps. We employ the Bidirectional Encoder Representations from Transformers (BERT) model to classify the parking sentiment and conduct regression analyses to explore its relationships with socio-spatial factors. Findings reveal significant variations in parking sentiment across POI types and CBSAs, with Restaurant POIs showing the most negative. Regression results further indicate that denser urban areas with higher proportions of African Americans and Hispanics and lower socioeconomic status are more likely to exhibit negative parking sentiment. Interestingly, an opposite relationship between parking supply and sentiment is observed, indicating increasing supply does not necessarily improve parking experiences. Finally, our textual analysis identifies keywords associated with positive or negative sentiments and highlights disparities between urban and rural areas. Overall, this study demonstrates the potential of a novel data source and methodological framework in measuring parking sentiment, offering valuable insights that help identify hyperlocal parking issues and guide targeted parking management strategies.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to quantify the public's perception of parking by analyzing crowdsourced online reviews, to identify the differences in parking perception between different regions and types of points of interest, and to explore the relationship between these differences and local socio - spatial factors. Specifically, the researchers hope to answer through the following three main questions:
1. **What are the differences in the public's perception of parking between different types of points of interest and Core - Based Statistical Areas (CBSAs)?**
The study reveals whether there are significant differences in people's attitudes towards parking by analyzing online reviews of different types of locations (such as restaurants, retail, entertainment venues, etc.) and different regions. For example, the study found that parking reviews for restaurants are the most negative, while reviews for personal services and entertainment venues are more positive.
2. **How are these differences related to local socio - spatial factors?**
Through regression analysis, the study explores the relationship between parking perception and factors such as population density, economic status, and ethnic composition. The results show that densely populated urban areas with higher population density, higher proportions of African - Americans and Hispanics, and lower socioeconomic status are more likely to exhibit negative parking perception.
3. **What are the reasons for different parking attitudes?**
By analyzing the content of review texts, the study identifies keywords related to positive or negative emotions, further revealing the specific reasons that affect parking perception. For example, increasing parking supply does not necessarily improve the parking experience, which may be because the quality and service level of parking facilities are equally important.
### Overview of Methods
To answer the above questions, the study adopts the following methods:
- **Data Preparation**: More than 4.98 million parking - related reviews were collected from Google Maps, covering 1.129 million points of interest (POIs), and the reviews were pre - processed. Natural language processing techniques (such as part - of - speech tagging) were used to screen out parking - related reviews.
- **Attitude Classification**: The BERT model was used to classify the reviews by sentiment, dividing them into four categories: positive, neutral, negative, and irrelevant, thereby extracting the public's attitude towards parking.
- **Socio - Spatial Regression Analysis**: By constructing a Generalized Additive Model (GAM), the relationship between parking perception and socioeconomic, demographic, and land development factors was analyzed to explore how these factors affect parking perception.
### Main Findings
- **Differences in Parking Perception by Type of Point of Interest**: Parking reviews for restaurants are the most negative, while reviews for personal services and entertainment venues are more positive.
- **Differences in Parking Perception by Region**: In large urban areas, especially in densely populated city centers, parking perception is more negative.
- **Impact of Socio - Spatial Factors**: High population density, low socioeconomic status, and higher proportions of African - Americans and Hispanics are associated with negative parking perception, while rural areas show more positive parking perception.
### Conclusion
This study demonstrates the potential of using new data sources (such as online user reviews) to evaluate the public's perception of parking, providing valuable insights for urban planners and local managers, and helping to identify parking problems in specific areas and develop targeted management strategies.