Spatial Mismatch Between Transportation Development and Tourism Spatial Vitality in Yunnan Province in the Context of Urban–Rural Integration

Abstract

As China’s urban–rural integration progresses, the connections between urban and rural areas continue to strengthen, making the spatial matching between transportation infrastructure and tourism resources increasingly crucial for coordinated regional development. This study investigates the spatial–temporal mismatch between transportation development and tourism spatial vitality in Yunnan Province, proposing optimization strategies to improve their coordination. Using Weibo check-in big data and OpenStreetMap transportation network data, we apply Convolutional Long Short-Term Memory (ConvLSTM) networks and bivariate spatial autocorrelation analysis to examine this relationship. The results show strong transportation–tourism matching in Kunming and surrounding areas. However, northwest and southern Yunnan exhibit significant mismatches—despite transportation improvements, underdeveloped tourism resources constrain vitality growth. Particularly in some remote regions, well-developed transportation infrastructure coexists with low tourism vitality, revealing persistent spatial mismatches between transport facilities and tourism resources. In general, transportation infrastructure development generally enhances tourism spatial vitality, but requires coordinated tourism resource development and market demand alignment. The study results provide a basis for improving the coordinated development of transportation and tourism, offering practical guidance for policymakers to promote balanced regional development and urban–rural integration.

1. Introduction

In recent years, China’s urban–rural integration has strengthened the connections between cities and rural areas, emphasizing the flow of resources and spatial connectivity. As important pathways for urban–rural integration, transportation and tourism development can promote regional collaboration and complement urban and rural resources [1,2]. However, a spatial mismatch persists between transportation infrastructure and the spatial distribution of tourism resources, hindering efficient mobility and limiting the potential for tourism development [3]. Therefore, studying the spatial matching relationship between transportation development and tourism spatial vitality can provide a scientific theoretical basis and practical guidance for advancing regional urban–rural integration.

While existing research has gradually gained academic attention, few studies deeply examine the dynamic changes in transportation–tourism spatial vitality mismatches, particularly within urban–rural integration contexts. Specifically, the current literature fails to fully reveal the interactive relationship between transportation development and tourism resource utilization, especially in quantifying their spatiotemporal mismatches. This study therefore addresses two key questions: (1) Does transportation–tourism mismatch exist in Yunnan at different time points? (2) How does this mismatch manifest spatially? We hypothesize that transportation infrastructure improvements will gradually reduce regional mismatches between transportation and tourism spatial vitality.

2. Literature Review

2.1. Urban Vitality and Tourism Spatial Vitality

Currently, research on the relationship between transportation development and tourism spatial vitality is gaining academic attention, but there are still some shortcomings and issues. From the perspective of tourism spatial vitality, existing studies mainly focus on the development of tourism resources and the analysis of tourist flows [4,5]. Although big data analysis methods have been widely applied in socioeconomic research in recent years [6], their integration with transportation and tourism studies remains insufficient. These studies have not fully utilized data resources such as traffic flow, tourist behavior trajectories, and transportation networks [7], thus failing to reveal the deep relationship between transportation and tourism spatial vitality [8,9]. Therefore, spatial analysis methods based on big data technology can more effectively uncover the dynamic impact of transportation development on tourism spatial vitality across time and space, providing more precise theoretical support and practical recommendations.

Research on regional spatial vitality can be traced back to Jane Jacobs’ theory of urban vitality. She argues that urban vitality stems from functional diversity and frequent human interactions, particularly when supported by convenient transportation and good spatial accessibility, which promote economic and social development. This theory establishes the foundation for regional spatial vitality studies [10,11]. Regional spatial vitality depends not only on spatial structure but also on the flow and agglomeration of factors like population, capital, and information, serving as a core indicator for measuring regional economic and social vitality [12]. Early studies primarily employed qualitative methods, focusing on functional zoning, transportation infrastructure, and economic agglomeration effects [13]. With the introduction of big data technology, research has gradually shifted toward quantitative approaches. Scholars now analyze population mobility and economic activity intensity using new data sources like mobile phone signaling data, social media data (including Weibo check-in data), and nighttime light data [14,15,16]. Spatial statistics and spatiotemporal behavior analysis tools have also been employed to reveal the spatiotemporal distribution patterns of spatial vitality. Current research employs various spatial analysis methods including spatial autocorrelation analysis, fractal theory, space syntax, and entropy analysis [17,18,19,20]. These quantitative approaches effectively uncover the complexity of regional spatial vitality and its spatiotemporal variations. However, while these methods demonstrate significant advantages in revealing dynamic changes of spatial vitality, they face certain limitations. For instance, current spatial vitality research relies heavily on big data, making analytical results highly dependent on data quality and accessibility [21,22].

2.2. Research on Big Data Applications in Tourism Spatial Vitality

Tourism spatial vitality differs from general spatial vitality by focusing specifically on tourist behaviors and destination dynamics. Early studies on tourism spatial vitality mainly used qualitative analysis, discussing the attractiveness of tourism destinations, the spatial distribution of tourists, and movement patterns [23,24]. These studies emphasized tourism resources, facility distribution, and transportation infrastructure development [25]. With the rapid advancement of big data technologies, recent studies have shifted toward quantitative analysis, utilizing more granular data sources to reveal dynamic characteristics of tourism spatial vitality [26]. Diverse datasets including GPS traces, mobile signaling data, social media data, and tourism big data have become crucial sources for investigating tourism spatial vitality [17,27,28,29]. By analyzing tourists’ travel trajectories, dwell times, and visiting paths, scholars can quantitatively assess the vitality levels of tourism areas and the concentration of tourists in different spatial blocks [30,31].

Recent studies on tourism spatial vitality increasingly rely on social media big data, particularly Weibo posts, travel blogs, and user reviews as primary research materials [32,33]. These social media data provide scholars with richer and more dynamic information on tourism activities, allowing for an in-depth analysis of tourist behavior patterns, the vitality of tourist regions, and the interaction between tourists and tourism resources [34]. Social media data, especially Weibo check-in data, have become an important tool for assessing tourism spatial vitality [35]. Researchers now use such data to examine tourists’ stay duration, visit frequency, and activity preferences at destinations [36]. For instance, Weibo check-in data help investigate the reasons behind tourists’ site selections and their spatial distribution patterns, allowing for the evaluation of vitality levels across different destinations [37,38]. Additionally, social media data reveal tourists’ interactions and emotional expressions during trips, offering valuable insights for tourism marketing and management [39]. By analyzing tourist dynamics on Weibo, scholars can also gain insights into tourists’ travel patterns, routes, and preferences for different tourism resources [40], providing a basis for more targeted tourism management and transportation planning. Beyond spatial distribution, Weibo check-in data combined with user-generated content (comments, photos, hashtags) enables researchers to understand tourists’ interests, preferences, and emotional tendencies [41,42,43]. These findings help uncover the intrinsic factors of tourism spatial vitality, particularly the relationship between visitor behavior and attraction appeal.

2.3. Study on the Mismatch Between Urban Transportation Development and Tourism Spatial Vitality

Recent years have witnessed growing academic interest in the relationship between urban development and tourism spatial vitality. Accelerated urbanization, with its expanding urban spaces and diversified functions, has drawn significant attention to the spatial distribution of tourism activities, tourist behaviors, and vitality levels across urban areas [44]. Researchers examine this relationship through multiple dimensions including urban spatial structure, transportation networks, and socioeconomic activities [45]. Some studies analyze functional zoning and transportation accessibility to reveal variations in tourism attractiveness and economic activity density across different urban areas [34]. Others focus on the integration of urban infrastructure with tourism resources, demonstrating that areas with better transportation access tend to attract more visitors and enhance local economic and social vitality [26]. In the context of urbanization, transportation development emerges as a key driver of urban spatial vitality. Transportation systems directly influence tourist flows and spatial accessibility while indirectly determining tourism vitality levels [46]. Particularly in large cities, the completeness of transportation networks, facility distribution, and route efficiency significantly affect tourist mobility and destination choices. Building on this understanding, scholars employ big data analytics and spatial statistics to investigate the interaction between transportation networks and tourism spatial vitality. These studies reveal how urban transportation development shapes the spatiotemporal distribution of tourism activities and visitor aggregation patterns [47,48].

Despite many studies revealing the interaction between transportation and tourism spatial vitality, the existing literature offers limited exploration of the mismatch between transportation development and tourism spatial vitality in the context of urban–rural integration [49]. This issue is particularly prominent in regions like Yunnan Province, which boasts abundant natural resources and unique cultural characteristics, where significant mismatches persist between transportation infrastructure construction and tourism resource distribution. To address this research gap, this study employs Weibo check-in big data analysis combined with neural network models to investigate the spatial mismatch between transportation development and tourism spatial vitality in Yunnan Province. Specifically, this study aims to (1) utilize Weibo check-in big data and OpenStreetMap transportation network data, employing Convolutional Long Short-Term Memory (ConvLSTM) networks and bivariate spatial autocorrelation analysis, to investigate the dynamic mismatch between transportation infrastructure and tourism spatial vitality in Yunnan Province; (2) analyze the coordination between transportation development and tourism resource utilization across different time periods, revealing the spatiotemporal evolution of transportation–tourism mismatches. Finally, this study proposes practical recommendations for enhancing urban–rural integration by improving the synergy between transportation development and tourism spatial vitality in Yunnan.

3. Materials and Methods

3.1. Study Area

Yunnan Province, located in China’s southwestern frontier (Figure 1), possesses some of the country’s richest tourism resources, making tourism one of its key pillar industries [50]. However, the province’s complex topography, dominated by plateau mountains, significantly impacts transportation infrastructure layout and construction difficulty. Many tourism-rich areas suffer from poor transportation accessibility, limiting the full utilization of tourism resources and constraining tourism spatial vitality. This inherent contradiction between terrain and transportation development makes Yunnan an ideal case for studying spatial matching relationships between transportation development and tourism spatial vitality, particularly for revealing spatial coordination challenges in geographically complex regions. In recent years, as Yunnan accelerates its digital transformation in tourism, many key tourist cities and attractions within the province have implemented systems for collecting and analyzing visitor behavior data. The accumulation of big data, such as Weibo check-in data and social media interactions, provides a solid data foundation for studying tourism spatial vitality. Consequently, Yunnan serves as an exemplary case for investigating spatial mismatch between transportation development and tourism spatial vitality, offering valuable theoretical and practical insights for other tourism-oriented regions with complex terrain and uneven urban–rural development.

3.2. Data Source

This study utilizes OpenStreetMap (OSM) as the primary source for transportation data, which offers advantages including strong openness, high update frequency, and detailed road information, making it widely applicable for transportation network and spatial accessibility research [51]. After obtaining Yunnan Province’s road network data through the OSM platform and third-party data services, we conduct data-cleaning and processing procedures. These include removing redundant and duplicate records, completing missing road segments, standardizing road type labels, and optimizing road topology structures to ensure data accuracy and usability. The cleaned OSM transportation data not only preserve the spatial characteristics of road networks but also maintain excellent compatibility for spatial analysis, establishing a solid foundation for subsequent spatial matching analysis between tourism spatial vitality and transportation networks. The study collects transportation data for three time points: 2013, 2018, and 2023 in the following Figure 2.

As one of China’s most widely used social media platforms, Weibo provides check-in data and geotagged content that contain abundant authentic and dynamic tourist activity information [42,52]. Compared with traditional tourism statistics, Weibo data demonstrate significant advantages including high frequency, strong timeliness, broad coverage, and relatively precise location accuracy, making them particularly valuable for tourism spatial vitality research. Weibo check-in data directly record users’ stays and activities at specific times and locations, accurately reflecting tourists’ spatial behavior patterns, travel preferences, and route choices. By analyzing large-scale check-in samples, researchers can identify tourist hotspots, peak periods, and visitor concentration levels across different spatial units, enabling the construction of spatiotemporal patterns of tourism spatial vitality. Moreover, Weibo data possess strong social interaction properties. In addition to location information, they also include user-generated content such as comments, images, and tags, which facilitate the analysis of tourists’ emotional evaluations and subjective experiences of tourist destinations, helping to assess tourism appeal and regional vitality levels.

This study collects Weibo check-in data through web-scraping techniques and third-party API interfaces, obtaining historical check-in records for Yunnan Province for the years 2013, 2018, and 2023 in the following Figure 3. The raw dataset contains multiple fields including user IDs, timestamps, geographic coordinates, location names, city tags, and content summaries, providing a comprehensive foundation for subsequent analysis. We implement a rigorous data-cleaning process with five key steps: First, check-in points are filtered by spatial range and time period to exclude irrelevant records. Second, abnormal users and commercial accounts are removed to eliminate interference. Third, the check-in coordinates are standardized and matched to administrative divisions, assigning the data to the corresponding counties or tourist attractions. Fourth, the data are spatially aggregated according to the research scale, with metrics such as check-in frequency and active user count calculated for each region. Finally, the data undergo normalization to eliminate the effects of population density and urban hierarchy differences on vitality measurement, ensuring that the Weibo data accurately reflect the tourism spatial vitality level of each region.

3.3. Methods

3.3.1. Convolutional Long Short-Term Memory Network (ConvLSTM)

Weibo check-in data demonstrate distinct spatiotemporal characteristics, where user check-in behaviors not only vary dynamically over time but also exhibit spatial clustering and diffusion patterns, effectively reflecting tourism spatial vitality [53]. Conventional static modeling approaches (e.g., regression analysis, factor analysis) struggle to adequately capture these spatiotemporal dynamics. The ConvLSTM network proves particularly suitable for addressing such challenges [54].

ConvLSTM is a deep learning model that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks. It captures long-term dependencies in time series data while preserving spatial structure. Unlike traditional LSTM, which uses fully connected operations, ConvLSTM replaces matrix multiplications in inputs, hidden states, and gating mechanisms with convolutional operations, maintaining spatial information and demonstrating superior performance for spatiotemporal sequence data processing. This approach proves particularly suitable for transforming Weibo check-in data into spatial grid images of time series for modeling. It efficiently extracts and predicts tourism vitality evolution patterns across different regions while revealing spatial correlations and vitality intensity variations between areas.

The computational formulas of ConvLSTM are as follows:

1. Input gate:

$$f_t=\sigma (W_{xi}\ast X_t+W_{hi}\ast H_{t-1}+b_i)$$

where $i_t$ represents the activation value of the input gate, controlling the degree of information flow. $\sigma$ is the sigmoid function, $\ast$ denotes the convolution operation, $X_t$ is the input of Weibo check-in data at the current time, and $H_{t-1}$ is the hidden state from the previous time step.

2. Forget gate:

$$f_t=\sigma (W_{xf}\ast X_t+W_{hf}\ast H_{t-1}+b_f)$$

where $f_t$ determines which information to discard.

3. Candidate memory cell:

$$C_t^{\prime }=\tanh (W_{xc}\ast X_t+W_{hc}\ast H_{t-1}+b_c)$$

where $C_t^{\prime }$ is the candidate memory cell at the current time and tanh is the hyperbolic tangent activation function, used to generate new memory content.

4. Memory cell update:

$$C_t=f_t\circ C_{t-1}+i_t\circ C_t^{\prime }$$

where $C_t$ is the updated memory cell at the current time, reflecting the updated memory content. This formula combines the effects of the forget gate $f_t$ and the input gate $f_t$.

5. Output gate:

$$o_t=\sigma (W_{xo}\ast X_t+W_{ho}\ast H_{t-1}+b_0)$$

where $o_t$ represents the activation value of the output gate, controlling the content output from the memory cell.

6. Final output:

$$H_t=o_t\circ {\tanh (C}_t)$$

where $H_t$ represents the output at the current time, reflecting the vitality information after time and spatial processing.

The ConvLSTM model employs optimization algorithms (e.g., gradient descent) to update network weights. It calculates loss functions (such as mean squared error or cross-entropy) and computes gradients through backpropagation to update weights for loss reduction. During training, the learning rate is set to 0.001, the batch size is 32, and 20 epochs are used with the Adam optimizer.

3.3.2. Bivariate Spatial Autocorrelation

Bivariate spatial autocorrelation is a statistical method used to analyze the spatial relationship between two variables [55]. Unlike traditional univariate spatial autocorrelation, bivariate spatial autocorrelation not only focuses on the spatial clustering characteristics of a single variable but also reveals the spatial linkage and mismatch between two different variables. This method proves particularly valuable for studying the relationship between transportation and tourism spatial vitality, as these systems interact but often develop asynchronously across regions. The bivariate spatial autocorrelation analysis effectively identifies spatial imbalances, such as areas with strong transportation development but weak tourism vitality. It serves as a powerful tool for assessing spatial coupling intensity and regional coordination, especially for logically related but potentially mismatched elements like transportation infrastructure and tourism spatial vitality [56]. Its primary advantage lies in quantitatively detecting spatial mismatch patterns, enabling the precise identification of underperforming areas.

$$I_{xy}={\displaystyle \frac{n}{{\sum }_i{\sum }_j{w}_{ij}}}·{\displaystyle \frac{{\sum }_i{\sum }_j{w}_{ij}(x_i-\overline{x})(y_j-\overline{y})}{\sqrt{\sum _i(x_i-\overline{x}{)}^2}·\sqrt{\sum _j(y_i-\overline{y}{)}^2}}}$$

where $x_i$ represents the transportation development level in a region, $y_i$ represents tourism spatial vitality, $\overline{x}$ and $\overline{y}$ are the average values of the two variables, $w_{ij}$ is the spatial weight matrix, indicating the neighboring relationship between regions $i$ and $j$, and $n$ is the total number of regions in the study area.

4. Results

4.1. Transportation Development in Yunnan Province

The transportation development in Yunnan Province for 2013, 2018, and 2023 is shown in Figure 4. Overall, transportation infrastructure shows consistent annual improvement, with particularly significant progress in urban road networks and intercity connections. During this period, accelerated urbanization drives substantial road construction in major cities, especially Kunming, where city centers develop more comprehensive transportation networks. Enhanced road construction in central urban areas effectively improves traffic capacity and alleviates congestion during peak hours. From 2013 to 2023, Yunnan witnesses remarkable advancements in intercity highways, railways, and airport infrastructure, particularly in high-grade highways and high-speed rail construction. Improved connections between Kunming and cities like Dali and Gaoligong Mountains reduce travel time and promote regional economic integration. Beyond the main arteries, road density increases substantially. As cities expand, the construction of smaller access roads and secondary roads beyond the main arteries has gradually been strengthened, especially in suburban and rural areas. These roads provide more convenient travel conditions for residents and businesses. Furthermore, the construction of these auxiliary roads not only helps improve local traffic flow but also effectively reduces urban traffic pressure, enhancing the overall efficiency of the transportation system.

Overall, Yunnan Province has achieved remarkable progress in transportation network development over the past decade. The continuous improvement in road density and connectivity, both within cities and between urban areas, has provided strong support for economic growth and population mobility.

4.2. Analysis of Tourism Spatial Vitality in Yunnan Province

The analysis of tourism spatial vitality in Yunnan Province reveals significant changes over the past decade (Figure 5). In 2013, the tourism spatial vitality in Yunnan is relatively low. On one hand, the infrastructure for tourism is not fully developed, and the exploitation and use of tourism resources are relatively slow. On the other hand, tourism activities in Yunnan are mainly concentrated in Kunming and the core areas of several key prefecture-level cities. As the provincial capital, Kunming has long been the transportation and economic center of the province, attracting a large number of tourists. Therefore, the tourism vitality in Kunming’s core areas, such as Dianchi Lake and Xishan, is relatively high. However, except for Kunming and a few core cities, other regions exhibit lower tourism spatial vitality, reflecting a trend of concentration in tourism resources and markets. In 2018, the tourism spatial vitality in Yunnan Province reaches its peak. During this period, the province makes significant progress in tourism infrastructure development, resource utilization, and market expansion. The accelerated integration between tourism and other industries—particularly transportation, accommodation, and cultural sectors—facilitate rapid development and efficient use of tourism resources. Numerous high-vitality tourism areas emerge across urban and surrounding regions, with notable improvements observed in Dali, Lijiang, Shangri-La, and Yuxi, attracting large numbers of visitors. This change also indicates that Yunnan’s tourism industry gradually expanded beyond Kunming and core cities, with a clear trend toward extending to other regions. In 2023, the tourism spatial vitality in Yunnan Province experiences a decline. While Kunming and its surrounding areas maintain high tourism vitality, other cities see a noticeable decrease. Kunming’s abundant natural landscapes and historical–cultural resources have long made it the primary tourist destination. Continuous infrastructure improvements further enhance its attractiveness, intensifying tourist concentration in this core region. Although other cities in Yunnan, such as Lijiang, Dali, and Shangri-La, have seen rapid growth in tourism markets, their tourism resources gradually become saturated, and competition in the tourism market intensified. Consequently, some regions struggle to sustain high tourism vitality levels, particularly in remote areas or locations with underdeveloped infrastructure where visitor flows diminish further led to lower tourism vitality. Additionally, the COVID-19 pandemic’s profound global impact on tourism also affects the tourism industry in Yunnan. While the province’s tourism industry shows gradual recovery, pandemic-related disruptions slow the rebound in certain markets, particularly secondary/tertiary cities and remote areas traditionally dependent on tourist volumes, contributing to the 2023 vitality decline.

4.3. Mismatch Analysis Between Transportation Development and Tourism Spatial Vitality in Yunnan Province

Based on the results of the bivariate spatial autocorrelation analysis from 2013 to 2023, the mismatch between transportation development and tourism spatial vitality in Yunnan Province presents different spatial distribution characteristics at various time points (Figure 6). In 2013, only Kunming and its surrounding areas show a concentration of high tourism vitality and high transportation development (HH clustering). As the provincial capital, Kunming benefits from well-developed transportation infrastructure and abundant tourism resources, attracting substantial tourist flows. At this time, this region shows strong spatial matching, where dense and efficient transportation networks effectively support tourism activities. However, some peripheral areas around Kunming exhibit low tourism vitality and high transportation development (LH clustering). Although these regions have relatively well-developed transportation facilities, such as highways or other transportation hubs, their tourism vitality remains low due to a lack of sufficient tourism resources or the ineffective development of the tourism market. This phenomenon suggests that well-developed transportation infrastructure does not necessarily promote tourism development directly, indicating a spatial mismatch between transportation resources and tourism resources.

In 2018, several regions including Honghe, Puer, Yuxi, Dali, and Lijiang join Kunming in showing HH clustering patterns. This development indicates improved tourism resource utilization and enhanced transportation infrastructure in these areas, leading to better spatial matching between tourism vitality and transport development that supports local tourism growth. However, LH clustering becomes more pronounced this year, particularly around Kunming’s periphery. These areas maintain strong transportation infrastructure but demonstrate low tourism vitality due to underdeveloped tourism resources or insufficient market demand. This situation reflects that improved transportation infrastructure does not directly translate into tourism growth, highlighting a significant spatial mismatch between transportation and tourism. Finally, in 2018, HL clustering is relatively rare, suggesting that in most regions with high tourism vitality, the construction of transportation infrastructure has also received corresponding attention.

In 2023, the HH clustering area around Kunming contracts slightly. While Kunming remains Yunnan’s primary tourism and transportation hub, slowing tourism growth due to market saturation and resource development bottlenecks have led to moderated tourism spatial vitality increases. The LH clustering area also shrinks in 2023. Although these regions maintain developed transportation systems, their improved tourism resource utilization helps reduce transport–tourism mismatches. However, some remote areas still show low tourism vitality despite transportation accessibility. In 2023, the distribution of HL clustering further decreases. Most high-vitality tourism areas benefit from corresponding transportation improvements. This trend demonstrates Yunnan’s success in coordinating transport infrastructure development with tourism resource utilization, effectively reducing spatial mismatches between these systems.

Despite significant improvements in Yunnan’s transportation infrastructure in recent years, tourism resource development and market expansion in some regions have failed to keep pace. This has led to a mismatch between transportation development and tourism resources, with some well-connected areas failing to attract enough visitors. In remote areas, although transportation facilities have improved, the lack of sufficient tourism appeal results in lower tourism vitality. As tourist preferences grow increasingly diverse, visitors increasingly favor destinations like Kunming, Dali, and Lijiang that combine transportation convenience with rich tourism resources. By contrast, other areas struggle to develop comparable attractiveness, leading to declining tourism spatial vitality. This demand-side imbalance further exacerbates the transportation–tourism mismatch across the province. The tourism market in Yunnan primarily concentrates on well-known attractions, such as Kunming, Dali, and Lijiang, where transportation is convenient and tourism facilities and services are relatively well-developed. However, most tourist attractions in Yunnan have not gained sufficient popularity or appeal, leading to a situation where, despite transportation coverage in these areas, the tourism spatial vitality cannot be fully realized due to the lack of sufficient market attraction and supporting tourism facilities. Tourists’ interests and demands are focused on a few popular destinations, while other regions experience low visitor flow and insufficient tourism vitality. Additionally, although transportation infrastructure in Yunnan has made significant progress, its coordination with the tourism industry still needs improvement. Transportation development tends to focus on roads, railway networks, and other hardware facilities, while lacking in-depth exploration of tourism resources, market demand forecasting, and the enhancement of related services. Even if transportation conditions improve, the lack of synchronized development in tourism services, cultural activities, and other support factors still prevents the effective enhancement of tourism spatial vitality.

5. Discussion

This study systematically analyzes the spatial mismatch between transportation development and tourism spatial vitality in Yunnan Province for 2013, 2018, and 2023, utilizing Weibo check-in big data and OpenStreetMap road network data with ConvLSTM and bivariate spatial autocorrelation analysis methods. The findings hold significant implications for promoting urban–rural integration. The identified spatial mismatches between transportation and tourism reveal imbalanced resource allocation and development opportunities between urban and rural areas. In some areas with developed transportation but insufficient tourism vitality, although improvements in transportation infrastructure provide conditions for regional development, the lack of effective tourism resource development and market expansion leads to significant differences in tourism vitality between urban and rural areas. Addressing this mismatch can both enhance transportation investment returns and foster economic interaction between urban and rural areas.

Many existing studies typically use traditional statistical data (such as tourist reception volume and tourism revenue) or survey-based methods to measure tourism vitality [57,58]. For example, some studies analyze tourism vitality in different regions of China through questionnaires and government statistical data, yet these approaches often lack the detailed capture of the real-time spatial distribution of tourist activities [59]. Other studies employ remote sensing and nighttime light data to estimate tourism vitality, which provide broad spatial coverage but struggle to distinguish tourist activities from other human behaviors and capture short-term or temporary tourism behaviors [59]. Compared to these methods, our use of Weibo check-in data offers distinct advantages in temporal immediacy and spatial precision. The platform records tourists’ social behaviors and movement trajectories in real time, enabling the accurate analysis of distribution patterns and activity hotspots during specific periods. This approach overcomes the temporal–spatial limitations of traditional statistics with their low collection frequency and coarse spatial scales [60]. Furthermore, Weibo check-in data capture tourist group behaviors more precisely than standalone remote sensing data or administrative statistics, providing more detailed spatial and behavioral-level insights.

By analyzing the spatial mismatch between transportation development and tourism spatial vitality in Yunnan Province, our study reveals the imbalance between transportation development and tourism vitality across different regions. Compared to other studies, our study emphasizes the dynamic mismatch between transportation and tourism vitality and provides specific details on spatiotemporal changes [61]. The findings demonstrate that Kunming consistently maintains high transportation–tourism matching, while other regions show varying degrees of mismatch—either having surplus transportation with inadequate tourism development or vibrant tourism with lagging transportation infrastructure. These results align with existing research findings [62]. For example, some studies indicate that transportation–tourism matching remains generally weak in China’s second-tier cities, particularly regarding internal resource allocation and facility distribution [48]. Our research further reveals that this mismatch in Yunnan Province demonstrates distinct temporal dynamics, showing an initial aggravation followed by gradual mitigation between 2013 and 2023, reflecting continuous adjustments in transportation–tourism interactions. Other studies have identified transportation–tourism mismatches across Chinese regions [63], but most focus on quantitative analysis and lack attention to the temporal evolution of the mismatch phenomenon [18]. In contrast, our study compares three time points (2013, 2018, and 2023) to demonstrate changing mismatch patterns in Yunnan, highlighting the infrastructure–tourism development relationship. For instance, accelerated transportation construction in 2018 helped Kunming and surrounding areas reach peak tourism vitality, while other regions failed to achieve synchronized growth, resulting in more pronounced spatial mismatches. Additionally, based on the conclusions drawn from the big data analysis, our analysis confirms that the development of transportation infrastructure plays a key role in enhancing tourism vitality. However, as some researchers note, even with relatively complete transportation infrastructure, many regions still experience transportation–tourism mismatches due to inadequate tourism resource development and market expansion [64,65]. Building on this, our study of Yunnan’s 2013–2023 transformation demonstrates that transportation improvements alone cannot guarantee tourism growth. Policy interventions, local economic development, and tourism resource exploitation equally contribute to achieving balanced development. The provincial case shows that these factors collectively determine the ultimate effectiveness of transportation investments in boosting tourism vitality.

This study employs Weibo check-in data to analyze the spatiotemporal distribution of tourism spatial vitality in Yunnan Province, revealing transportation–tourism mismatch patterns through big data analytics. However, the representativeness of Weibo data warrants careful consideration. While offering high temporal resolution and spatial precision for tracking real-time tourist behaviors and movement patterns, Weibo’s user demographics exhibit significant biases. First, Weibo users predominantly comprise younger, urbanized populations with frequent internet usage, potentially limiting the data’s representativeness for certain regions and demographic groups. This bias may particularly affect remote areas and older age groups, failing to fully capture comprehensive tourist behavior patterns [66]. Second, these inherent biases may distort evaluations of niche tourism destinations, particularly those less favored by younger or tech-savvy travelers. To better understand the limitations of Weibo check-in data, this study compares its representativeness with other tourism data sources. Official tourism statistics, typically collected by government agencies or travel companies, provide comprehensive coverage of tourist numbers and flows. However, these datasets often suffer from temporal–spatial lags and may be constrained by their collection methodologies and coverage limitations. Alternative data sources like Ctrip’s user reviews offer real-time spatial coverage, yet they also exhibit user preference biases and depend on voluntary contributions, potentially overlooking offline tourist populations. To validate Weibo data’s representativeness, we recommend cross-verifying it with other datasets to assess consistency in reflecting vitality patterns across major attractions and remote areas. Future research could incorporate more diversified data sources, including tourist expenditure records and travel agency itineraries. Such multi-dimensional data integration would compensate for Weibo data’s representational gaps and enable more comprehensive analyses.

This study identifies a spatial mismatch between tourism vitality and transportation development in Yunnan’s remote areas, particularly resource-rich regions in northwest and southern Yunnan, where transport infrastructure lags behind. To advance urban–rural integration, it is essential to increase investment in transportation infrastructure in these areas, particularly strengthening the connectivity between rural and urban areas and improving accessibility in remote regions. By improving the construction of highways and railway networks, regional connectivity can be enhanced, thereby stimulating local tourism vitality and promoting the joint development of urban and rural economies. Moreover, transportation infrastructure improvements should not rely solely on physical construction, but must integrate with local tourism resources through targeted planning and development. Governments and relevant agencies should formulate customized transportation plans based on regional tourism characteristics to prevent transportation–tourism mismatches. Finally, regional coordination development policies should be established, especially for economically weaker areas, providing financial support and policy incentives to promote the joint development of the tourism industry and transportation infrastructure.

6. Conclusions

This study examines the spatial matching relationship between transportation development and tourism spatial vitality in Yunnan Province by integrating Weibo check-in data with OpenStreetMap transportation data, revealing significant mismatches between infrastructure and tourism activity. The results demonstrate distinct regional variations in Yunnan’s transportation–tourism matching. Kunming and its surrounding areas maintain high levels of matching, serving as the province’s tourism vitality core. However, northwest and southern Yunnan exhibit different patterns—despite gradual transportation improvements, underdeveloped tourism resources constrain full vitality potential in these regions. Overall, our findings demonstrate that transportation improvements significantly enhance destination accessibility, enabling easier tourist access to remote attractions and boosting regional tourism vitality. The development of high-grade highways, railways, and airports particularly reduces travel time between destinations, facilitates tourist flows, and stimulates tourism growth. However, the impact of transportation development on tourism spatial vitality requires reasonable development based on the specific tourism resources of each area to further match these factors and promote urban–rural integration.

Based on our research findings, this study proposes the following practical recommendations to promote coordinated development between transportation and tourism: First, increase investment in transportation infrastructure in remote areas. The government should prioritize regions with relatively underdeveloped transportation systems such as northwest Yunnan and southern Yunnan, particularly by enhancing connectivity with core cities to reduce transportation barriers and facilitate urban–rural integration. Second, synchronize tourism resource development with transportation infrastructure construction. While improved transportation creates conditions for tourism development, without simultaneous tourism resource development and market expansion, transportation convenience cannot be fully transformed into tourism vitality. Therefore, it is recommended to strengthen tourism resource exploration and development alongside transportation infrastructure projects, especially in remote areas with untapped tourism potential. Third, optimize the diversified layout of tourism products. To reduce the concentration of tourism resources, local governments should be encouraged to develop diverse tourism products based on local characteristics to attract different types of tourists. For example, cultural and ecotourism products could be developed in northwest Yunnan and southern Yunnan by leveraging their unique natural landscapes and cultural resources to enhance regional tourism appeal. Fourth, strengthen the integration of tourism with local economies. Tourism serves not only as an engine for economic growth but also as an important driver for urban–rural integration. Policymakers should encourage local governments to integrate tourism with other industries such as agriculture, culture, and environmental protection to generate broader positive economic impacts. Fifth, establish long-term monitoring and dynamic adjustment mechanisms. Given the spatiotemporal dynamics of transportation–tourism spatial vitality mismatches, the government should create a long-term monitoring system for coordinated transportation–tourism development and make timely policy adjustments based on monitoring results to ensure more balanced resource allocation during the urban–rural integration process.