Hotspot Identification and Causal Analysis of Chinese Rural Tourism at Different Spatial and Temporal Scales Based on Tourism Big Data

Abstract

Rural tourism serves as a crucial means for fostering rural economic prosperity and inheriting rural culture. The assessment of the quality of rural tourism development and the identification of disparities in rural tourism development among regions have become focal points in current research. This paper utilizes tourism big data to establish a system for evaluating rural tourism popularity and proposes a method for identifying rural tourism hotspots. The study explores the spatiotemporal evolution characteristics and formation mechanisms of the cold and hot patterns of rural tourism in China during two periods (pre-pandemic and post-pandemic) and on two spatial scales (provincial and municipal levels). The research findings indicate that (1) the annual variation in rural tourism popularity exhibits a fluctuating upward trend, with significant seasonal variations on a monthly basis. (2) The spatial pattern of rural tourism popularity changes with the scale effect. At the provincial level, hotspot areas form an east–west dual-core pattern, while at the municipal level, hotspot areas demonstrate an evolution from a three-core to a four-core pattern. In the post-pandemic era, rural tourism popularity in the northwest and southwest regions is experiencing a counter-trend growth. (3) At different spatiotemporal scales, influencing factors and their impact intensities vary. At the provincial level, road density and reception capacity consistently play dominant roles, and per capita disposable income significantly influences early-stage popularity enhancement. At the municipal level, resident population and tourism resources influence are the dominant factors, and the influence of air quality and regional media attention gradually strengthens. This article provides a new perspective on quantitative research in rural tourism, offering significant guidance for the rational allocation of resources in rural tourism, regional tourism collaboration, and the sustainable development of rural tourism in the post-pandemic era.

1. Introduction

In China, the countryside, with its beautiful natural environment and unique local culture, is an essential carrier for people to return to nature and remember their nostalgia; therefore, it is increasingly favored by tourists. At present, rural tourism has become an important market segment of the domestic tourism economy, and both short-distance tours and rural tours have become the primary forms of tourism and leisure alternatives to inter-provincial medium- and long-distance tours [1]. Especially since the introduction of the rural revitalization strategy, rural tourism has become a new attempt to extend the tourism industry to traditional agriculture. Rural tourism is of great practical importance to optimize the agrarian industrial structure and promote the recovery of the rural economy and the in-depth implementation of rural revitalization [2]. At the same time, with the development and popularization of the Internet, tourism websites, social software, and networking platforms have become essential tools for tourists to query travel information. These tools assist in travel decision-making and sharing travel experiences, and the resulting tourism big data provide a vast data source and methodological support for rural tourism research. Therefore, under the new situation of the rising status of rural tourism, the research on the spatio-temporal evolution characteristics of rural tourism and its influencing factors based on network big data has important practical value. The obtained knowledge can be used for the rational development and allocation of rural tourism resources and for guiding the tourism management department to implement scientific and reasonable decisions and arrangements.

Rural tourism research has traditionally received notable attention from the academic community as an interdisciplinary and multi-paradigm research field. The research perspective of tourism geography is diversified and informative, involving tourism, geography, sociology, and other disciplines. Research on rural tourist sites from the perspective of tourism geography mainly focuses on the aspects of spatial agglomeration patterns and evolutionary characteristics of rural tourist sites. Early research studies mainly focused on the concept and characterization of the spatial agglomeration pattern of rural tourist sites. Farms, small towns, and other settlements are the main subjects of study [3]. Location theory and “core-edge” theory have also been applied to the spatial identification of village and town settlements [4,5]. With an increasingly abundant variety of rural tourism types, such as agritainment, rural B&Bs, and ecological health, the spatial form, location scale, and distribution characteristics of rural tourism sites have received much attention from the academic community. However, most existing studies have relied on micro-scale explorations of smaller villages’ social and spatial characteristics, and macro-scale research needs to be strengthened. From the spatial and temporal evolution characteristics perspective, the study mainly focuses on the spatial differentiation pattern and spatial trajectory optimization of rural tourist spots [6,7]. The static layout of the spatial evolution process has been studied in depth, while research on the dynamic evolution remains underexplored [8,9]. In summary, most existing studies are micro-static analyses of specific rural tourism sites. Therefore, research on meso-macro scale rural tourism sites needs to be improved. In addition, in the current scenario, in which many rural tourism spots have a wide distribution range and low scale rank, it is difficult for government departments to obtain macro-dynamic data over a long time series. So, data acquisition has become a challenge for rural tourism research on the meso-macro scale.

The arrival of the big data era provides more possibilities for rural tourism research. At the beginning of the 21st century, Ettredge et al. [10] and Cooper et al. [11] applied web search data to study unemployment rates in the USA and cancer-related topics, respectively. Since then, network data analytics has been widely applied in social, economic, and tourism fields. Many tourism geographers have started using Google Trends and Baidu Index to study tourism. Xiang and Pan [12], Park et al. [13], Padhi and Pati [14], Önder [15], as well as Siliverstovs and Wochner [16] have monitored the size of the tourism industry, selected tourism destinations, and forecasted tourism demand using Google Trends and Twitter data. In China, the Baidu Index has been widely used for tourism network attention research, mainly focusing on spatial and temporal distribution [17,18,19,20], passenger flow predictions [21,22], and influencing factors [19,23]. Internet celebrity tourist sites [24], ancient city tourism [18], cave tourism [19], ice and snow tourism [25], urban sports tourism [26], and heritage tourism [27] have also garnered attention. In addition, the potential of attractions, tourist behavior, and emotional perception have been studied using geotagged photographs and network texts. Wood et al. [28], García-Palomares et al. [29], Giglio et al. [30], and Paldino et al. [31] have used geotagged photographs to detect the potential of tourist attractions, tourist behavior, and the attractiveness of city residents in major cities around the globe. Clavé [32], Kim [33], and Marine-Roig [34] have examined destination image perceptions in Barcelona, Paris, and Athens, respectively, using online travel reviews; Philander and Zhong explored tourist emotions in integrated resorts using network texts [35]. Previous research has mainly focused on micro-analyses of specific tourist sites while lacking explorations of the spatial and temporal evolution characteristics and relevant influencing factors of rural tourism at meso and macro multi-scales. Regarding data acquisition, scholars have primarily relied on single-network platforms, such as Google Trends or the Baidu Index, and the analysis of influencing factors has been dominated mainly by descriptive statistics and correlation analyses. For this reason, this paper takes 31 provinces and 367 prefectural-level cities in China as the research object. It explores the method of measuring rural tourism popularity based on tourism big data. On this foundation, two time periods before and after the epidemic were selected temporally, and two scales of provinces and prefectural cities were used spatially to analyze the hot and cold patterns and evolutionary characteristics of rural tourism popularity in China. Finally, the causes of the evolution of the patterns of provinces and prefectural cities were analyzed with the help of a geodetector. The research explored the spatiotemporal heterogeneity law of spatial and temporal distribution and influencing factors of rural tourism popularity on different scales. Additionally, the analysis of rural tourism popularity across different spatial scales reveals the presence of scale effects in rural tourism popularity. The obtained data provide theoretical guidance for the rational development of rural tourism resources and the sustainable development of post-pandemic rural tourism.

2. Materials and Methods

2.1. Data Sources

Three main types of data were used in this paper: (1) network data, which were mainly used to evaluate rural tourism popularity, including four major types of network data—social data, long video data, short video data, and online tourism data; (2) statistical data, which were mainly obtained from national, provincial, and municipal statistical yearbooks and statistical bulletins; and (3) vector data. which were used for ArcGIS spatial analysis and visualization. These data were downloaded from the Yangtze River Delta Science Data Center.

2.2. Methods

2.2.1. Research Framework

Following the research framework (Figure 1) of ‘Rural Tourism Popularity Measurement-Temporal Cold Hot Spot Identification-Spatial Cold Hot Spot Identification-Causes Analysis,’ this paper establishes a Rural Tourism Popularity Evaluation Indicator System based on the three dimensions of social media, online videos, and online travelogues. It assesses rural tourism popularity in China from 2013 to 2022. Building upon this foundation, the study utilizes the seasonal intensity index and spatial autocorrelation methods to explore the spatiotemporal evolution characteristics of the cold- and hot-spot pattern of rural tourism in China across various spatial and temporal scales. Finally, by employing a geodetector model, the research uncovers the influencing factors and mechanisms behind the spatiotemporal evolution at the provincial and prefecture levels.

2.2.2. Construction of Evaluation Indicator System

In this paper, an evaluation index system (Table 1) was constructed from three dimensions for the Chinese rural tourism popularity. These dimensions were online video, social media, and online travelogue. Specifically, the following four first-level indicators were included: the network social index, long video index, short video index, and network travelogue index. The eleven secondary-level indicators were Weibo, WeChat, BiliBili, Tencent Video, Iqiyi, Youku, Tiktok, Kuaishou, Lumama, Tongcheng, and Ctrip. All indicators are dimensionless, and the hierarchical analysis method was used to determine weights. A total of 28 experts were asked to score their judgment, and the average value was taken to obtain the evaluation judgment matrix. The maximum characteristic root of the matrix λ_max = 4.0815, the consistency of the judgment matrix CI = 0.0272, the average stochastic consistency index RI = 0.89, and CR = 0.0305 < 0.1. The data passed the consistency test, and this method was used to derive secondary indicator weights.

The social network index includes data from Weibo and WeChat. During the study period from 1 January 2013 to 1 January 2023, 210,478 valid Weibo data were collected, and the number of likes was utilized as a measure of rural tourism popularity. WeChat had accumulated 5106 articles since 2015, and the number of reads was employed as a measure. The top four long video websites in terms of number of active people were selected: BiliBili, Tencent Video, Iqiyi, and Youku. Using “rural tourism” as the search term, the number of video plays and the heat index were used as the data source. The final search resulted in a total of 3348 videos for the period of 2021–2022. The short video index was selected from the top two short video apps (TikTok and Kuaishou). Using “rural tourism” as the search term and the number of likes of short videos as the data source. Finally, 2966 short videos were obtained from TikTok for 2021–2022 and 474 were obtained from Kuaishou for 2022. Lumama, Tongcheng, and Ctrip were selected as the three most renowned Chinese online tourism websites. By selecting “rural tourism” as the search term, the number of travelogue views was used as the rural tourism measurement of rural tourism popularity standard. Finally, travelogue data of 2934, 7432, and 5323 were obtained for Lumama, Tongcheng, and Ctrip, respectively.

2.2.3. Coefficient of Elasticity

In the context of the rapid development of the Internet, the elasticity coefficient is used to measure the relationship between rural tourism popularity and the number of Internet users to further explore the annual change in rural tourism popularity, and the formula is as follows [36]:

(1)$ET=\frac{∆G/G}{∆N/N}$

where ET is the elasticity coefficient; ΔG and G are the incremental and total amount of rural tourism popularity; ΔN and N are the incremental and total amount of Internet users); ET > 1, ET = 1, and ET < 1, which represent that the growth rate of rural tourism popularity is more significant than, synchronized with, and smaller than the growth rate of Internet users, respectively.

2.2.4. Seasonal Intensity Index

Tourism activities are characterized by temporal imbalance, often referred to as seasonality, and whether rural tourism is also subject to seasonality needs to be further verified. In addition, the outbreak of COVID-19 will also impact the time distribution characteristics of rural tourism. Typically, the concentration of tourism demand’s temporal distribution is caused by seasonality, measurable through the seasonal intensity index S, and the formula is as follows [27]:

(2)$S=\sqrt{\frac{\sum _{i=1}^{12}{{(M}_i-8.33)}^2}{12}}$

where S is the seasonal intensity index; $M_i$ is the ratio of the monthly rural tourism popularity to the annual total rural tourism popularity. The larger S is, the more concentrated it is in a certain period of time, and the smaller S is, the more dispersed it is.

2.2.5. Spatial Autocorrelation (Moran’s I)

In order to further explore the spatial distribution characteristics of rural tourism popularity, this paper utilizes spatial autocorrelation techniques to analyze the hot and cold patterns of rural tourism popularity and reveal the spatial distribution patterns and interrelationships of rural tourism popularity so as to gain insights into the development process of rural tourism popularity and explore the spatial dependence between geographic phenomena. Commonly used indicators are global and local spatial autocorrelation, Moran’s I [37]. The former measures the overall degree of clustering of observables within a certain space, while the latter measures the degree of clustering of observables among units within a spatial context:

(3)$I=\frac{n}{\sum _i\sum _j{w}_{ij}}\times \frac{\sum _i\sum _j{w}_{ij}\left(x_i-\overline{x}\right)\times \left(x_j-\overline{x}\right)}{\sum _i{\left(x_i-\overline{x}\right)}^2}$

(4)$I_i=\frac{n^2}{\sum _i\sum _j{w}_{ij}}\times \frac{\left(x_i-\overline{x}\right)\sum _j{w}_{ij}\left(x_j-\overline{x}\right)}{\sum _j{\left(x_j-\overline{x}\right)}^2}$

where I and $I_i$ represent global and local spatial autocorrelation Moran’s I, respectively; i and j denote the provinces (including autonomous regions and municipalities) or prefectural-level cities; and $w_{ij}$ is the spatial weight matrix. In this paper, the method of Queen’s adjacency [38] was adopted to define the spatial weight matrix.

2.2.6. Geodetector

As a spatial statistical method, geodetector, with its factor detection and interaction detection, can measure factor effects and interactions. By utilizing this method to calculate the q-value and interaction value of each influencing factor, the geodetector can explain the spatial variability of the dependent variable and make inferential statistics on its distribution pattern. This study utilized geodetectors to explore the extent to which the influencing factors themselves and their interactions explain the spatial and temporal heterogeneity of rural tourism popularity [39,40]. The formula is:

(5)$q=1-\frac{\sum _{H=1}^L{N}_h{\sigma }_h^2}{N{\sigma }^2}$

In Equation (5), q is the influence of factor X on rural tourism; N is the number of samples in the study area; ${\sigma }^2$ and ${\sigma }_n^2$ are the variances for all regions and subregions, respectively; and L is the number of subregions. If ${\sigma }^2\ne 0$, the model holds true. The value of q is [0, 1], and the larger the value of q, the stronger the influence of factor X on rural tourism popularity and vice versa. The interaction detector was employed to investigate the combined impact of different influencing factors (Table 2). When the influencing factor X1 interacts with X2, it exerts influence on the dependent variable “popularity of rural tourism” in five different ways: Weaken, uni-; Weaken, nonlinear-; Enhanced, bi-; Enhanced, nonlinear-; and Independent.

2.2.7. Selection and Calculation of Influencing Factors

Most scholars believe that the economy [41], resources [42], population [43], and service capacity [44] will affect the spatial distribution of tourism areas. As a new form of tourism occurring in rural areas, rural tourism has the typical characteristics of traditional tourism projects, and its evolution in tourism’s spatial pattern is influenced by the spatial layout of traditional tourist attractions. Still, rural tourism has its own unique features. Consequently, the research selects 18 and 10 indicators from economic development, tourism resource endowment, ecological and environmental conditions, population, transportation conditions, service reception, and the level of Internet development, respectively, as candidate variables to study the impact of rural tourism popularity at the provincial level and prefectural municipal level. Following the correlation test, it was revealed that five variables at the provincial level and six variables at the prefecture level exhibited a high correlation with rural tourism popularity (Table 3).

The five variables that affect rural tourism popularity in provinces are as follows. (1) Per capita disposable income: Per capita disposable income is proportional to the standard of living. An increase in per capita disposable income will drive the tourism market toward a period of demand growth. (2) Population density: Population reflects the size of the local tourism source market. Considering the significant area differences, population density was chosen as a variable at the provincial level. (3) Road density: Highway transportation serves as the primary mode of transportation for rural tourism. The density of highways directly influences the accessibility of rural tourist attractions, thereby affecting tourists’ willingness to travel. (4) Tourism reception capacity: This indicator is the sum of the number of star-rated hotels and travel agencies. More travel agencies and star-rated hotels can promote the development of high-quality tourism services through market competition and stimulate potential tourism motivation through word-of-mouth effects [45]. (5) Internet development level: The level of Internet development affects the number and willingness of tourists to post rural tourism experiences on the Internet. This paper uses the Internet access rate as a measurement indicator.

The six variables affecting rural tourism popularity in prefecture-level cities are as follows. (1) Tourism revenue: Tourism revenue is the source of the expanded reproduction of tourism, providing financial guarantees for the optimal allocation of tourism resources and the realization of the benign development of tourism. (2) Tertiary industry value added: According to the theory of the stage of development of the three industries, the growth of the tertiary industry’s value-added represents the transformation of the high end of the industrial structure, enhancing the quality of the tourism support system and improving the comprehensive experience of rural tourism and other industries [46]. (3) Influence of rural tourism resources: Tourism resources are the core and carrier of rural tourism development. From a Baidu index search, five hundred twenty-six typical rural tourism scenic spots were selected from 4807 tourist attractions above the 4A level. Following the established keyword search criteria in existing research [27], scenic spots not entered into the Baidu index were excluded, resulting in a total of 237 scenic spot samples. The scenic spot Baidu index was used to determine the influence of rural tourism resources in each region. (4) Air quality: Ecological conditions have become a major factor affecting tourism demand and preference. Considering data accessibility, this paper selected the air quality excellence rate to characterize the impact of air quality on rural tourism popularity. (5) Resident population: At the municipal level, the difference in the area between cities is small, so the total resident population was selected to reflect the market size. (6) Regional media attention index: This index reflects the regional capabilities in rural tourism network marketing and indirectly signifies the local emphasis on rural tourism. The attention index utilizes “rural tourism” as the key term. Data retrieval from Bing search news websites is conducted using Python. The news media release platforms are categorized into national, provincial, and local levels. National media assigns a value of 1 point to each prefectural-level city, Provincial media assigns 0.5 points to prefecture-level cities in the province, and prefectural-level media assigns 0.2 points to the respective prefectural-level city. The final rural tourism media attention index is thus derived.

3. Results

3.1. Time Evolution Characteristics

3.1.1. Inter-Annual Variability

Examining the data from past years, it is evident (Figure 2a) that the trend of rural tourism popularity can be divided into four phases: steady growth, explosive growth, a brief decline, and rapid rebound in the post-epidemic era. From 2013 to 2015, a stable growth stage characterized rural tourism popularity as relatively low but in a positively growing phase with fluctuating growth rates. Moving to 2015–2019, an explosive growth stage unfolded, with a more significant increase observed, except in 2018. This noteworthy growth is attributed to the introduction of policies such as the ‘Guiding Opinions on the Development of Leisure Agriculture’ and the ‘Guidelines and Opinions on Promoting Sustainable Development of Rural Tourism’. These policies explicitly defined the development objectives, main tasks, and safeguard measures for rural tourism development, providing essential guidance and practical significance for its promotion. In 2019–2020, rural tourism popularity appeared to experience a brief weakening. The outbreak of the COVID-19 pandemic in some areas of the country toward the end of 2019 led to concerns among the masses about participating in outdoor activities. To some extent, this also affected the attention and participation levels in rural tourism. From 2020 to 2022, the popularity of rural tourism experienced a rapid resurgence. Despite lingering uncertainties due to the pandemic during this period, the increasing refinement of epidemic prevention policies and the public’s desire for rural outdoor activities following lockdowns have contributed to a growth in rural tourism popularity. To gain a deeper understanding of the changes in rural tourism popularity, in conjunction with the fluctuations in the number of netizens, a comparative analysis utilizing the elasticity coefficient reveals a triple-peaked pattern (Figure 2b). In most years, the elasticity coefficient exceeded 1, signifying that the growth rate of rural tourism popularity outpaces that of netizens. However, the elasticity coefficient was lower in 2015, 2018, and 2020. The decline in the elasticity coefficient is attributed to the initial introduction of the “Internet+” action plan during the Twelfth National People’s Congress in 2015. With more traditional industries embracing the Internet through online-to-offline (O2O) initiatives, the widespread adoption of the Internet has significantly increased, leading to a decrease in the elasticity coefficient. The year 2018 marked the golden period for comprehensive tourism development. The flourishing growth of ice and snow tourism, heritage tourism, and other aspects somewhat diverted attention from rural tourism. Additionally, the outbreak of the COVID-19 pandemic in 2020 diminished residents’ inclination to travel outdoors.

3.1.2. Inter-Month Variations

Figure 3a shows the monthly rural tourism popularity as well as the percentage of the year. Monthly changes in rural tourism popularity are generally characterized by “double peaks and three valleys,” with May and October constituting “double peaks” and January, June, and December forming “three valleys”. Overall, in April, May, July, August, September, and October, rural tourism popularity was higher, representing the peak season of rural tourism, while in January, February, and December, rural tourism popularity was lower, meaning the off-season of rural tourism. In April and May, when the spring blossoms are in full bloom, coupled with the Qingming and May Day vacations, the number of people willing to take trips increases significantly. During the summer months of July and August, student groups and parent–child tours become the main customers of rural tourism. September and October are the best time for picking fruits and vegetables and enjoying the autumn scenery, and during the National Day holiday, the number of tourists increases significantly. Therefore, rural tourism popularity shows more apparent seasonal fluctuations. The seasonal intensity index in Figure 3b indicates that all index values exceed 3. This suggests that while monthly variations in rural tourism popularity are apparent, the yearly changes are not prominent, and the status of the differences is relatively stable. In 2020, the index value peaked at 4.14, primarily attributed to the outbreak of the COVID-19 pandemic, resulting in a sudden decline in rural tourism popularity in the early months of the year. By contrast, 2021 witnessed the lowest index value, at 3.02. This decline is primarily a result of the gradual recovery of rural tourism popularity in the post-epidemic period and the increased enthusiasm for short-distance rural tourism among the populace, contributing to a relatively balanced distribution of the month’s characteristics throughout the year.

3.2. Provincial Spatial Scales

3.2.1. Evolution of the Spatial Structure

The spatial autocorrelation of rural tourism popularity was examined using the global Moran’s I index. The Moran’s I index is positive and fluctuates between 0.255 and 0.411, indicating a solid spatial autocorrelation of rural tourism popularity. Additionally, when applying the localized Moran’s I index to analyze the rural tourism popularity of 31 provinces in 2013, 2016, 2019, and 2022, respectively (Figure 4), it was found that over the past ten years, China’s rural tourism popularity in the high–high aggregation zone (hot spot zone) exhibited an east–west dual-core structure and demonstrates an evolutionary trend of gradual diffusion from north to south. On the other hand, the low–low aggregation area (cold spot area) underwent a mono-dual nucleus evolutionary process, with the cold spot area in the northwest gradually shrinking in scope and the northeast becoming a new cold spot area. In 2013, the hot spot area for rural tourism in the east was concentrated in the Yangtze River Delta region and the Fujian–Jiangxi region, while the hot spot area in the west was located in the Chengdu–Chongqing region. The cold spot area was centrally distributed in Xinjiang, Tibet, Qinghai, and other northwestern inland provinces. By 2016, the hotspot area had further expanded to the southeast, and the scope of the cold spot area gradually narrowed. In 2019, the western hotspot area extended to include Yunnan Province. In 2022, the Northeastern region became a new cold spot area. This resulted in the formation of the eastern hot spot area centered on the Yangtze River Delta and the Fujian and Jiangxi regions, the western hot spot area centered on the Chengdu–Chongqing region, and two cold spot areas in the northwest and northeast regions. Evidently, a robust trickle-down effect was present in hotspot areas, and the collaborative interaction between regions profoundly impacted the optimization of the spatial structure of rural tourism. Despite boasting a rich and diversified rural regional culture, the northwest region faces challenges, including a relatively fragile ecological environment, the rough development of rural tourism products, and weak economic support. In the northeast region, despite having a reasonably complete infrastructure, the vitality of tourism is gradually weakening due to the slowdown in economic and population growth.

3.2.2. Evolution of Provincial Patterns

To understand the detailed changes in the rural tourism popularity of each province, the rank order of each province was subdivided into three tiers: 1–10 indicated hot spots, 11–20 indicated general areas, and 21–31 indicated cold spots. Changes in these tiers can also be categorized into three types: stabilized type, ascending type, and descending type (Figure 5). Specifically, most of the eastern provinces were in the hotspot area, especially the Jiangsu and Zhejiang Provinces, which always remained among the top four. Shandong, Hebei, and Hainan Provinces were in the general area. Tianjin Municipality was in the ascending type, which improved from the cold spot area to the general area, with its ranking increasing from 24 to 17. Most of the central provinces had a wide distribution and showed unstable changes. Anhui and Jiangxi always belonged to hot spot areas, and their rank order jumped between 4 and 8. Hunan ranged between 15 and 16 and belonged to the general area; the rank order of Hubei was always 21, identifying it as belonging to the cold spot area; Shanxi and Henan belonged to the descending type, both descending from a general area to a cold spot area. Most of the western provinces were in the cold spot area, and their rank order remained relatively stable. However, Sichuan and Chongqing were in the hot spot area; Inner Mongolia, Shanxi, Guangxi, and Guizhou were in the general area; Gansu, Qinghai, Ningxia, Xinjiang, and Tibet ranged between 23 and 30, were in the cold spot area, and their rank order was relatively stable; Yunnan followed an increasing trend, increasing from the general area to the hot spot area. Finally, all three provinces in the northeast were found to be cold spot regions with decreasing rankings.

3.3. Spatial Scale of Prefecture-Level Cities

3.3.1. Evolution of the Spatial Structure

At varying administrative scales, the spatial pattern of rural tourism hotspots exhibits inconsistencies influenced by scale and zoning effects. Within the administrative structure of China, prefecture-level cities function as pivotal entities bridging higher and lower administrative levels, constituting critical players in regional tourism development. Therefore, researching and analyzing the popularity of rural tourism in Chinese prefecture-level cities can facilitate the discernment of development patterns and trends across different scales and zones. This endeavor holds indicative value for precise positioning and targeted measures at a more refined scale.

Overall, from 2013 to 2022, the global Moran’s I index of rural tourism popularity was positive and fluctuated between 0.201 and 0.462, indicating spatial solid autocorrelation at the prefecture-level city scale. Additionally, the local autocorrelation of rural tourism popularity in the four typical years of 2013, 2016, 2019, and 2022 was analyzed. The results (Figure 6) revealed apparent differences in rural tourism popularity among prefectural-level cities, with a significant clustering feature in geospatial location. The high–high aggregation area of rural tourism popularity (hotspot area) gradually transitioned from the triple nucleus to the quadruple nucleus, and the low and low aggregation area (cold spot area) shifted from the northwestern inland area to the northeastern area. This differs from the results at the provincial level, indicating a certain degree of scale effect in the study of rural tourism hot and cold patterns.

In 2013, rural tourism hotspots were distributed in the Beijing–Tianjin–Hebei region, Chengdu–Chongqing area, and the border area of Jiangsu–Zhejiang–Anhui–Fujian–Jiangxi, forming three hotspot regions in the northern, southwestern, and southeastern parts. These regions were primarily concentrated in core cities such as Beijing, Tianjin, Chongqing, Chengdu, Shanghai, Shangrao, Huangshan, Suzhou, Hangzhou, and Huzhou. The main contributing factors were the relatively developed economies in the northern and southwestern hotspot areas, characterized by a high degree of rural tourism industrialization and scale, driven by a robust demand for short-distance rural “micro-tourism” [47]. The southeastern hotspot region emerged as a core area for rural tourism development, leveraging its advantages in tourism resource endowment and well-developed supporting facilities. Representative regional rural cultural elements, such as the traditional customs of drying crops in the autumn sunshine in Wuyuan, the Huizhou culture in Yixian County, and the homestay culture in Moganshan, became the foundation of its early advantages. Cold spot areas were predominantly located in the northwestern inland regions, where weak regional economic support and extensive development of tourism products hindered rural tourism development in the northwest. In 2016, the southeastern hotspot broke the segmentation and isolation of rural tourism resources, driving the surrounding areas of Nanping, Fuzhou, Xiamen, and other cities to form a ring-core extension group. The spillover effect in the tourism space was significant. The cold spot range in the northwest gradually narrowed, with Altay and Hulunbuir standing out as emerging cities for rural tourism development, leveraging their border landscapes and grassland culture. By 2019, the hotspot area continued to expand, with the southwestern region having the highest proportion of characteristic villages of China’s ethnic minorities, accounting for as much as 44.67% of the national total. With the increasing enthusiasm for minority cultures, the hotspot area in the southwestern region expanded to cities such as Aba, Ganzi, and Qiandongnan. The cold spot region in the northwest was limited to certain cities in Xinjiang. In 2022, in the post-pandemic era, rural tourism with ethnic minority characteristics in the northwestern borderlands and southwestern China achieved counter-trend growth. Yunnan’s Dali, Lijiang, Chuxiong, and Kunming formed the fourth hotspot region. The cold spot areas shifted from the northwestern region to the northeastern region.

3.3.2. Evolution of the Prefecture-Level City Pattern

LISA’s local spatial autocorrelation analysis unveils the overall distribution pattern of rural tourism popularity at a certain level. However, it cannot accurately discern each research unit’s hot and cold change status. Therefore, referring to relevant research findings, we employed ArcGIS 10.2 software to classify the rural tourism popularity rankings in 2013 and 2022 into ten hotness levels using the equal interval classification method. Subsequently, spatial distribution maps of rural tourism popularity rankings in 2013 (Figure 7a) and 2022 (Figure 7b) were generated. Additionally, a spatial evolution map of rural tourism hot and cold grades was created on the basis of the changes in hot and cold grades (Figure 6c). This map was further categorized into three types: stable, ascending, and descending, aiming to accurately identify changes in the hot and cold patterns of rural tourism. The subsequent analysis of the results is presented as follows.

Overall, rural tourism popularity exhibited a decreasing trend from southeast to northwest between 2013 and 2022. The majority of hotspot areas were concentrated southeast of the Hu Huanyong Line. Over the course of a decade, the overall pattern of rural tourism popularity did not undergo fundamental changes. From 2013 to 2022, a total of 92 cities experienced changes in rural tourism popularity in hot and cold grades, with an overall change rate of 25.1%. Cities without changes in popularity grades dominated, and the spatial transfer of rural tourism popularity grades demonstrated varying degrees of path-locking.

Examining Figure 7c, it is evident that ascending cities can be categorized into three types: western inland cities, eastern coastal cities, and satellite cities around large urban centers. Western inland ascending cities, including Altay, Hami, Turpan, Haixi, Linzhi, and others, exhibit an ascent greater than 50. These Western inland cities have the advantage of rural tourism resource endowment, characterized by unique natural landscapes and local customs. However, their rural tourism development is significantly constrained by the level of economic development and tourism marketing strategy in the Western region. With the improvement of transportation conditions and the popularity of the Internet, inland cities in the West combine Western culture and Tibetan culture with Gobi agriculture, mountain recreation, desert cross-country, and other special rural tourism projects to shape the brand image of special regional rural tourism. Notable destinations like Kanas Lake, Qinghai Lake, and the Dunhuang Mogao Caves are emerging as new “Instagram-Worthy” rural tourism spots. Additionally, in the post-epidemic era, the northwest region, with its sparse population and low population density, adapted to the psychology of tourists seeking to avoid congestion. Simultaneously, in previous years, National Day was often the peak of outbound tourism. Under the impact of the global epidemic, most domestic tourists now choose long-distance domestic alternative tourism products or projects, further enhancing the rural tourism popularity of the Northwest.

Ascending-type cities in the eastern coastal region, such as Jiaxing, Nantong, Chizhou, Taizhou, and Lishui, exhibit a high degree of featured rural agglomeration and relatively superior natural geography. They have driven the development of the rural tourism industry in the entire region by relying on solid rural tourism-supporting facilities in the local area, coupled with an excellent level of development in regional tourism collaboration. The ascending-type satellite cities around large cities are mainly Zhangjiakou, Baoding, Langfang, Aba, Ganzi, and Qiandongnan. Among them, Zhangjiakou, Baoding, and Langfang are located in the vicinity of the Beijing–Tianjin–Tangshan urban agglomeration, which has formed a 1–2 h rural self-driving tourism circle with their mother cities. Aba, Ganzi, and Qiandongnan are situated around the Chengdu–Chongqing urban agglomeration, forming a southwest minority ecological rural tourism industry cluster by virtue of their unique minority regional characteristics and neighboring source markets.

From the spatial distribution of descending-type cities, it can be observed that these cities are primarily located in the northeast region, with some cities in the central region and the Hainan region. In the northeast region, descending-type cities like Mudanjiang, Hegang, Chaoyang, Huludao, Dandong, and others are situated. This region, being an old industrial base, has experienced a slowdown and even retrogression in economic development in recent years, leading to a significant population exodus. Consequently, the tourism market has contracted to varying degrees, and rural tourism’s popularity continues to decline. Central region’s descending-type cities include Luliang, Linfen, Heze, Zhoukou, Hengshui, and others. The fragile ecological environment of rural tourism and a single economic structure focused on mineral resource development pose challenges to transformation due to the depletion of resources. In the Hainan region, the decline is mainly attributed to the epidemic’s impact in cities like Sanya, affecting the overall passenger flow. Additionally, in recent years, the rising cost of tourism in the Hainan region and negative tourism news, such as “rip-offs,” have dampened the enthusiasm for tourism.

3.4. Empirical Analysis of Influencing Factors of Rural Tourism Popularity

3.4.1. Results and Interactions of Impact Factor Probes at the Provincial Level Probes

The value of rural tourism popularity in each province served as the dependent variable, with five provincial detection factors employed as independent variables. Given GeoDetector’s proficiency in analyzing categorical variables, the factor variables underwent discretization through the K-means classification algorithm. Subsequently, the p-values of the factors for the years 2013, 2016, 2019, and 2022 were calculated using GeoDetector (Excel) software, and the results are presented in Table 4.

Examining a single factor, it becomes evident that road density (RDEN) holds the highest explanatory power. The p-values of the four typical years are all greater than 0.5 and show a tendency to increase gradually. There is an apparent positive correlation between transportation accessibility and rural tourism popularity, as reflected by RDEN. This is consistent with the findings of other studies [48]. Notably, regions with high rural tourism popularity, such as Beijing, Shanghai, Shandong, Jiangsu, and Zhejiang, consistently rank among the top eight in terms of highway density. Improved transportation conditions can reduce the distance friction effect in tourism travel, and changes in transportation networks and innovations in transportation technology will contribute to the evolution of the spatial structure of rural tourism; especially, the improvement of the highway network is of great significance in the development of rural tourism, which is mainly based on self-driving tours.

Population density (PDEN) significantly positively affects rural tourism popularity, with the p-value for the four typical years consistently exceeding 0.3. PDEN serves as a reflection of the potential tourism market size in the region where the rural tourist site is situated, and a higher density implies a larger potential tourism market size. However, over time, from 2013 to 2022, the p-value decreased from 0.412 to 0.323, indicating that the impact of PDEN on rural tourism popularity is weakening. The reason may be due to the improvement of transportation conditions to shorten the journey time, and tourists have gained more space for tourism trips, so the proportion of out-of-towners’ sources is increasing.

From 2013 to 2022, the influence ranking of reception service capacity (RECA) has been at the top. It shows an increasing trend, indicating that regional RECA is essential for securing rural tourism sources. The quality and service level of tourism service facilities, represented by travel agencies and star-rated hotels, determine the perceived experience of tourists. This is particularly evident in the era of experience and network economies, where travelers share their tourism perceptions and experiences through platforms such as tourism websites and personal social media. These shared experiences influence potential tourists’ preferences, consequently impacting passengers’ actual flow.

The influence of the per capita disposable income (PCDI) on rural tourism popularity decreased from 0.402 in 2013 to 0.214 in 2022, suggesting the presence of an income ‘inflection point’ in rural tourism travel behavior choices. When residents’ income reaches a certain threshold, the impact of economic variables gradually diminishes. In situations in which residents’ income is limited, tourism decision-making heavily relies on economic factors. The higher the income level of the region, the greater the likelihood of residents participating in tourism. However, as society develops and income levels improve, people’s concept of tourism undergoes changes. Travelers’ decision-making becomes more flexible and independent, no longer relying solely on economic income.

The Internet development level (INDL) has been on an upward trend in terms of influence from 2013 to 2022, with the total consumption of online tourism in China reaching a trillion in 2021. The “Internet Plus initiative” has emerged as a new scene for mass tourism and a fresh source of momentum for intelligent tourism. People are increasingly inclined to share their rural tourism experiences online, and regional governments are giving more attention to the construction of tourism network platforms, further boosting rural tourism popularity.

As can be seen from Figure 8, the interaction shows Enhanced, bi- and Enhanced, nonlinear-, and the multi-factor interaction has a greater effect than the single-factor interaction, which indicates that the evolution of rural tourism popularity hot–cold pattern is influenced by multiple factors. In 2013, the interaction effect of PCDI∩RDEN (q = 0.856) was the strongest, followed by RDEN∩PDEN (q = 0.842), and relatively weak effects of other factor interactions, indicating that PCDI, RDEN, and PDEN are the main influencing factors in the initial period. From 2016 to 2022, the RDEN∩RECA interactions are all the strongest, indicating that in the later period, RDEN∩RECA is the most critical interaction influence factor for the evolution of the hot and cold pattern of rural tourism popularity. Finally, the influence of INDL, in combination with other factors, has also increased in the past decade.

3.4.2. Results and Interactions of Impact Factor Probes at the Prefecture-Level Probes

There are significant differences in the distribution of rural tourism resources within each province in China, and the influencing factors are also affected by scale effects. To some extent, provincial-scale studies can hide the detailed characteristics of the spatial units at the sub-level. Combining the detection of influencing factors at the prefectural and municipal scales will help identify the development pattern and influencing mechanism of rural tourism at a finer scale, which will be helpful for the government to implement precise positioning and targeted measures for rural tourism in various regions.

The results (Table 5) show that the factor with the most significant explanatory power is the total resident population (REPO). Rural tourism is characterized by a short in-transit time and a high revisit rate. The REPO of prefecture-level cities provides a large number of continuous potential sources of customers for local rural tourism scenic spots. In addition, during the epidemic period, residents’ traveling distance was limited to a certain extent, and most could only choose peripheral and local tours. Consequently, local rural tourism became the leisure choice of many urban consumers in the context of the epidemic. However, between 2013 and 2022, there is a declining trend in the influence of the REPO. This indicates that with the improvement of transportation conditions and the increasing popularity of scenic spots, the overall radius of rural tourism source areas is expanding.

Tourism income (TINC) and the added value of the tertiary industry (TEVA) have essentially equal explanatory power for the popularity of rural tourism. TINC and TEVA influence rural tourism development through the supply-side pathway. However, it is noteworthy that the impact of TINC is confined to the tourism system rather than the entire prefectural-level city or regional system. TINC constitutes the fundamental source of reinvestment in tourism, providing financial assurance for the reconstruction of local rural tourism and the maintenance of tourism facilities. Rural tourism destinations serve as concentrated areas for the tertiary industry, fulfilling the essential functions of a tourist destination while also concentrating industries such as catering, accommodation, and entertainment. TEVA is influenced by regional rural tourism development and, conversely, affects the popularity of regional rural tourism.

Tourism resource influence represents the quality of regional rural tourism resource endowment. At the prefectural-level city level, there is significant variation in rural tourism resources, and the richness and grade of tourism resources directly determine the pattern of rural tourism industry development in prefectural-level cities. For example, Moganshan, by utilizing hundreds of vacation villas and inns left by predecessors, has laid the foundation for the development of rural tourism in Moganshan. However, with the elevation of tourism levels and the emergence of new types of tourism products, the impact of tourism resource endowment is gradually weakening. In the case of Yuanjia Village in Shaanxi, where resource conditions are not advantageous, the comprehensive exploration of the local rich cultural advantages of Guanzhong folklore has transformed it into a folk-themed rural tourism destination, with significant brand effects in Shaanxi and even nationally.

Air quality (AIRQ) reflects the excellent quality of the ecological environment, which serves as a crucial attraction and the foundational basis for the sustainable development of rural tourism. The significance of ecological environment impact began manifesting in 2016, and its influence demonstrates a continuously strengthening trend. With the advent of the era of great health, a favorable ecological environment has become one of the determining factors in selecting rural tourism destinations. Conversely, a poorer ecological environment inevitably leads to potential tourist loss, adversely affecting the rural tourism market. This holds certain reference value for the relevant trends and goal decision-making in the development of rural tourism.

Finally, while regional media attention (RMEA) has a relatively lower impact on rural tourism popularity, its overall influence shows an upward trend. As China enters the post-epidemic period, local governments and tourism bureaus have chosen to provide tourists with online tourism travel services through the construction of online platforms for tourism networks and to meet the consumers’ needs for higher quality and safety of tourism through the innovation and upgrading of tourism products and services, such as virtual tourism, reservation tours, and neighborhood tours.

From the perspective of interaction factor scores (Figure 9), the interaction manifests as both dual-factor enhancement and nonlinear augmentation, indicating that the evolution of the hot and cold pattern of rural tourism popularity in prefectural-level cities is influenced by six factors collectively. From 2013 to 2022, the interaction of TRIN∩REPO is the strongest, consistently exceeding 0.93. Furthermore, the interaction values of TRIN and REPO with other influencing factors are significantly higher than those of other influencing factors, indicating that TRIN and REPO are the primary influencing factors affecting the hot and cold pattern of rural tourism in prefectural-level cities. Additionally, in the past decade, the interactions of AIRQ and REMA with other influencing factors have continuously strengthened, suggesting a gradual rise in the influence of these factors.

4. Discussion

4.1. The Impact of Rural Tourism Policies on Rural Tourism Popularity

Because of the limited access to rural tourism network data, rural tourism popularity data before 2013 have not been incorporated. The impact of iconic events on rural tourism popularity, such as the resolutions of the Fifth Plenary Session of the 16th Communist Party of China Central Committee in 2005, which proposed the construction of a new socialist countryside and the national rural tourism development program in 2009 and outlined China’s rural poverty reduction and development program in 2011 [49] and the measures therein, has yet to be explored. Given the more significant positive correlation between rural tourism popularity and passenger flow, this can be translated into exploring iconic events and policies’ impact on passenger flow. The growth rate of passenger flow was used to create Figure 10, and iconic events and policies affecting the development of rural tourism were selected. As can be seen in the figure, iconic events and policies and the lag effect they cause substantially impact rural tourism income, further corroborating that iconic historical events can attract substantial attention flows for rural tourism development. Therefore, it is essential to accurately portray the spatio-temporal connection between tourism events and tourism elements. This helps explore the scientific problems and development laws of rural tourism; promote the further deepening of rural tourism research; and improve the planning, decision-making, and development level of rural tourism in the new era.

4.2. Response to Previous Studies

At the methodological level, the identification of cold- and hotspots in rural tourism based on tourism big data aims to provide new research perspectives for rural tourism. In other tourism fields, scholars such as Gao et al. [27], Hoffmann et al. [50], and Li et al. [51] have also evaluated the attention to heritage tourism, sustainable tourism, and the vitality of 5A scenic spots by utilizing a multi-indicator system, such as tourism websites and social media. However, it should be emphasized that this paper further advances and improves the evaluation system and measurement method of tourism heat based on previous studies. The objective is to expand the depth and breadth of quantitative research on rural tourism and to provide ideas and methods for studying spatial and temporal patterns of rural tourism. Additionally, at the spatial scale, Wan et al. [20] conducted a study on the temporal and spatial distribution of rural tourism popularity based on the Baidu Index at the provincial level. Building upon this, this paper carries out a study on the identification of hot and cold spots of rural tourism popularity in China at the provincial and prefectural level, which makes up for the lack of quantitative research on China’s rural tourism popularity at the mesoscale and reveals the scale effects of the spatial pattern and the influencing factors of rural tourism popularity in different dimensions. It also aids in recognizing the development level and potential of rural tourism and its formation mechanism on multiple scales, providing methodological references for identifying lagging and unbalanced areas of rural tourism development in a more detailed way. Finally, in the study’s time scale, Solazzo et al. [52] and Pramana et al. [53] evaluated Italian and Indonesian tourism before and after the epidemic, proposing recovery countermeasures. This paper refers to the above studies and conducts a study on the change in rural tourism popularity before and after the epidemic, which will provide a certain reference value for the recovery and sustainable development of the rural tourism industry after the epidemic.

4.3. Revelations and Recommendations

The study on the identification of cold and hot spots of rural tourism shows that rural tourism’s popularity in the northwest region is increasing, but it still faces developmental challenges. Northwest China boasts rich natural resources and historical and cultural heritage; however, villages with unique resources and heritage require further exploration. Some regions also grapple with issues such as a relatively fragile natural ecological environment and insufficient economic and tourism facilities support. In the future development of rural tourism, the western region should enhance external transportation and tourism facilities, fully explore the region’s unique natural landscape and local customs, and establish distinctive rural tourism activities. Simultaneously, studying influencing factors indicates that the Internet plays an increasingly vital role in tourism. Therefore, future efforts should actively leverage virtual tourism, Internet live broadcasting, and other network promotion and marketing channels to shape the characteristics of the rural tourism brand and cultivate Western “Internet celebrity” rural tourism destinations [50].

Rural tourism popularity continues to decrease in most cities in the Northeast and some resource cities due to economic recession and population loss. The natural resources of the northeast region have unique advantages, and rural tourism can be integrated with ice and snow tourism, green tourism, and summer vacation tourism to develop local characteristics of rural tourism projects and build an all-round, multi-level special rural tourism brand system. In addition, the Fifth Plenary Session of the 19th Central Committee put forward the revitalization of the northeast as the goal of China’s 14th Five-Year Plan. The northeast should seize the historical opportunities for the development of rural revival and other national strategies to promote the rural green recycling industry rural tourism economy, and actively complete the transformation of the industrial structure.

Despite the counter-trend surge in rural tourism popularity during the pandemic in the northwest and southwest regions of China, driven by rural tourism with ethnic minority characteristics, rural tourism in most parts of China has been significantly impacted by the serious consequences of the COVID-19 pandemic. In the post-epidemic era, the total tourism demand has been further suppressed, with tourists mainly traveling locally and within the province. The tourism industry has gradually evolved from solving basic tourism needs such as sightseeing tours to creating comprehensive resources such as cultural products, comfortable spaces, and high-quality services to satisfy people’s needs for cultural aesthetics, idyllic life, and quality experiences, which has also provided opportunities for the sustainable development of rural tourism [51]. Therefore, the government-led diverse entities should persist in deepening top-level design, closely aligning with the post-epidemic era’s tourism consumer demands and changing the past exploitative development strategy for tourists on the basis of people-oriented and holistic tourism sustainable development principles. This meets the consumption demands of the middle class to venture into the countryside, experience the local culture of the tourism destination, and harmoniously interact with residents, and it constructs an ecosystem to meet the needs of a safe and livable rural tourism destination, thereby enhancing the tourism service and guaranteeing function of the destination city. This, in turn, relies on the radiation and diffusion effect of the metropolitan area to drive the development of rural tourism in itself and its neighboring small- and medium-sized satellite cities [52].

5. Conclusions

This paper focuses on rural tourism as the research object and explores the method of measuring rural tourism heat on the basis of tourism big data. On this foundation, two time periods before and after the epidemic are selected temporally, and two scales of provinces and prefectural cities are used spatially to analyze the hot and cold patterns and evolutionary characteristics of rural tourism popularity in China. Finally, the causes of the evolution of the patterns of provinces and prefectural cities are analyzed with the help of a geodetector. The results of the study show that:

• (1). Rural tourism popularity in 2013–2022 was generally fluctuating and rising, with higher heat in April, May, July, August, September, and October, which is the peak season of rural tourism, and lower heat in January, February, and December, which is the off-season of rural tourism. It is worth noting that the COVID-19 pandemic outbreak in January 2020 led to a sudden drop in rural tourism popularity. With the gradual improvement of epidemic prevention and control measures, the heat in 2021 returned to the level before the epidemic.

• (2). The spatial pattern of cold and hotspots in rural tourism varies with scale and zoning effects. At the provincial level, there are significant regional differences in the popularity of rural tourism, with hotspots exhibiting a dual-core structure in the east and west and showing a diffusion trend from north to south. Cold spot areas have undergone an evolutionary process from a “single core” to a “dual-core” structure. At the prefectural city level, the overall pattern manifests as a “higher in the south, lower in the north, higher in the east, and lower in the west” spatial characteristic. The evolution pattern of rural tourism hotspots at this scale shows a “three-core to four-core” pattern, while cold spot areas shifted from the northwestern region to the northeastern region. In the post-epidemic era, the border regions of northwest and southwest China have witnessed a counter-trend growth in rural tourism popularity driven by rural tourism with ethnic minority characteristics.

• (3). At the provincial and prefectural city levels, the cold and hot pattern of rural tourism popularity is jointly influenced by multiple factors. However, there are differences in the specific influencing factors, and the intensity of their impact varies across different periods. At the provincial level, transportation conditions have consistently been the dominant factor, with factors influencing the early and mid-term popularity of rural tourism being more related to per capita disposable income and population density. The level of reception services and the development of the Internet significantly impact the later-stage popularity of rural tourism. Population size and resource endowment are the main factors affecting rural tourism popularity at the prefecture-level city scale. Air quality and regional media indexes have a smaller level of influence initially, but the positive impact is increasing.

The tourism big data for public service contains attribute information such as tourism time, place, and perception, characterized by rich and real-time data. On the basis of this, our study constructs a rural tourism popularity evaluation system, providing a new research path for quantitative studies in rural tourism. Additionally, the article explores the spatial and temporal evolution and influencing factors of rural tourism hot and cold patterns from two spatial perspectives, provincial and municipal, and two periods before and after the epidemic. Theoretically, it addresses the lack of quantitative research on rural tourism in China at the mesoscale, complementing the study and application of scale effects in tourism geography. Practically, the study recognizes the development level and potential of rural tourism and its formation mechanism at multiple scales, providing a methodological reference for identifying the lagging and unbalanced areas of rural tourism development in a more detailed way. Furthermore, the comparative analysis of the pre- and post-epidemic periods and the suggestions made will provide a particular value of reference for the recovery and development of the rural tourism industry after the epidemic.

However, constrained by the limitations imposed by data acquisition, more data from the consumer side of rural tourism were selected. Detailed rural tourism production and construction end data can be used for a more in-depth exploration of the mechanisms underlying rural tourism development, thus realizing rural tourism’s digital development and governance. The digital wisdom scenic area construction is also a future direction of rural tourism development. At the same time, while constructing this evaluation index system, certain network platforms have missing data for the year indexes. It may be possible that the existing indexes can be further enriched and perfected in the future as the network platforms continue to develop. Finally, this paper takes “rural tourism” as the keyword, and while this choice is scientific and universal, whether it covers all relevant forms of attention of travelers to rural tourism needs to be verified.

Agricultural production is a production process highly integrated with natural conditions [54]. The challenge and focus of future rural tourism development lie in how to introduce digitization in rural development, leverage the precise information flow provided by big data and online platforms, and further strengthen the positioning, observation, and empirical research of typical rural tourism destinations through the innovation of rural tourism theories, methods, and technologies.

Author Contributions

Y.F. wrote this article; Y.F. completed the data collection and processing work; C.F. and Z.C. reviewed the whole paper and put forward suggestions for improvement. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Acknowledgments

The authors would like to acknowledge all the reviewers and editors.

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. The research framework of this paper.

Figure 2. (a) Annual count of rural tourism popularity from 2013 to 2022; (b) elasticity coefficient of rural tourism popularity from 2013 to 2022.

Figure 3. (a) Monthly count of rural tourism popularity from 2013 to 2022; (b) seasonal intensity index of rural tourism popularity from 2013 to 2022.

Figure 4. Spatial distribution of rural tourism popularity by province across China in (a) 2013, (b) 2016, (c) 2019, and (d) 2022.

Figure 5. Change in the order of rural tourism popularity in 31 provinces for 2013–2022.

Figure 6. Trends in the temporal and spatial distribution of rural tourism popularity in prefecture-level cities, (a) 2013, (b) 2016, (c) 2019, and (d) 2022.

Figure 7. Spatial distribution of rural tourism popularity rankings in 2013 (a) and 2022 (b); (c) Spatial evolution of hot and cold ranks in rural tourism from 2013 to 2022.

Figure 8. The results of interaction detection at the provincial scale in 2013, 2016, 2019, and 2022, Note: Band * is Enhanced, bi-, while Band + is Enhanced, nonlinear.

Figure 9. The results of interaction detection at the prefecture scale in 2013, 2016, 2019 and 2022. Note: Band * is Enhanced, bi-, while Band + is Enhanced, nonlinear.

Figure 10. China’s 2013–2022 rural tourism visitor annual growth rate.

Table 1

Rural tourism popularity evaluation index system.

Level Indicators	Weights	Level 2 Indicators	Weights	Data Time Frame/Year	Metrological Standard
Network Social	0.3533	Weibo	0.2587	2013–2022	Number of likes
		WeChat	0.0946	2015–2022	Number of views
Short Video	0.1570	Tiktok	0.1149	2021–2022	Number of likes
		Kuaishou	0.0421	2022	Number of likes
Long Video	0.2541	BiliBili	0.0724	2021–2022	Number of plays
		Tencent	0.0643	2021–2022	Number of plays
		Iqiyi	0.0597	2021–2022	Number of plays
		Youku	0.0578	2021–2022	Number of plays
NetworkTravelogue	0.2356	Lumama	0.0538	2016–2022	Number of views
		Tongcheng	0.1033	2013–2022	Number of views
		Ctrip	0.0786	2013–2022	Number of views

Table 2

Types of interaction between two covariates.

Interaction Detection	Expressions	Meaning
Weaken, uni-	q(X1 ∩ X2) < min(q(X1), q(X2))	Indicates that the nonlinearity weakened after the mutual interaction of X1 and X2.
Weaken, nonlinear-	min(q(X1), q(X2)) < q(X1 ∩ X2)< max(q(X1), q(X2))	Indicates that the monoclinic line weakened after the mutual interaction of X1 and X2
Enhanced, bi-	q(X1 ∩ X2) > max(q(X1), q(X2))	Indicates that X1 and X2 enhanced each other after mutual interaction
Enhanced, nonlinear-	q(X1 ∩ X2) > q(X1) +q(X2)	Means that nonlinearity is strengthened after the mutual interaction of X1 and X2
Independent	q(X1 ∩ X2) = q(X1) +q(X2)	Suggests that X1 and X2 are independent of each other

Table 3

Selection and definition of influencing factors.

Variable	Meaning of Variables	Abbreviation	Level
Per capita disposable income	Income available for discretionary spending	PCDI	Provincial
Road density	Total road length-to-area ratio	RDEN	Provincial
Population density	Permanent population-to-area ratio	PDEN	Provincial
Reception capacity	Sum of travel agencies and star-rated hotels	RECA	Provincial
Internet development level	Internet access rate	INDL	Provincial
Tourism income	Domestic tourism income	TINC	Prefecture
Tertiary value added	Value added of the tertiary sector in a year	TEVA	Prefecture
Tourism resource influence	Network attention of rural tourism Sites	TRIN	Prefecture
Resident population	Number of resident populations at the end of the year	REPO	Prefecture
Air quality	Air excellence rate	AIRQ	Prefecture
Regional media attention	Media attention index	RMEA	Prefecture

Table 4

Results of the factor detection analysis at the provincial scale in 2013, 2016, 2019, and 2022.

Influencing Factor	PCDI	RDEN	PDEN	RECA	INDL
2013	0.402 ***	0.538 ***	0.412 ***	0.304 **	0.187
2016	0.232 **	0.581 ***	0.471 **	0.322 **	0.236 *
2019	0.236 **	0.625 **	0.418 **	0.377 **	0.122 **
2022	0.214 *	0.663 **	0.323 **	0.339 **	0.217 **

Note: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively.

Table 5

Results of the factor detection analysis at the municipal scale in 2013, 2016, 2019, and 2022.

Influencing Factor	TINC	TEVA	TRIN	REPO	AIRQ	RMEA
2013	0.428 ***	0.327 ***	0.542 ***	0.716 **	0.087	0.398 **
2016	0.332 **	0.337 **	0.522 ***	0.713 **	0.092 **	0.372 **
2019	0.366 **	0.309 **	0.414 **	0.692 **	0.115 **	0.412 **
2022	0.371 **	0.311 **	0.459 **	0.694 **	0.134 **	0.422 **

Note: ***, ** denote significance at the 1%, 5% levels, respectively.

References

1. Lu, L.; Li, L.; Hou, Y. Resilience and high-quality development of tourism destinations under the epidemic crisis. Tour. Trib.; 2022; 37, pp. 1-3. [DOI: https://dx.doi.org/10.19765/j.cnki.1002-5006.2022.09.001]

2. Liu, Y. Research on the urban-rural integration and rural revitalization in the new era in China. Acta Geogr. Sin.; 2018; 73, pp. 637-650. [DOI: https://dx.doi.org/10.11821/dlxb201804004]

3. Christaller, W. Some considerations of tourism location in Europe: The peripheral regions-under-developed countries-recreation areas. Pap. Reg. Sci. Assoc.; 1964; 12, pp. 95-105. [DOI: https://dx.doi.org/10.1007/BF01941243]

4. Weaver, D.B. Peripheries of the periphery. Ann. Tour. Res.; 1998; 25, pp. 292-313. [DOI: https://dx.doi.org/10.1016/S0160-7383(97)00094-7]

5. Britton, S.G. The Spatial Organisation of Tourism in A Nec Colonial Economy: A Fiji Case Study. Pac. Viewp.; 1980; 21, pp. 144-165. [DOI: https://dx.doi.org/10.1111/apv.212003]

6. Go, F.M.; Govers, R. Integrated quality management for tourist destinations: A European perspective on achieving competitiveness. Tour. Manag.; 2000; 21, pp. 79-88. [DOI: https://dx.doi.org/10.1016/S0261-5177(99)00098-9]

7. Pearce, D.G. Tourism in Paris Studies at the Microscale. Ann. Tour. Res.; 1999; 26, pp. 77-97. [DOI: https://dx.doi.org/10.1016/S0160-7383(98)00051-6]

8. Pavlovich, K. The evolution and transformation of a tourism destination network: The Waitomo Caves, New Zealand. Tour. Manag.; 2003; 24, pp. 203-216. [DOI: https://dx.doi.org/10.1016/S0261-5177(02)00056-0]

9. Li, Q.; Ma, X.; Shen, Y. Analysis of spatial pattern of rural settlements in northern Jiangsu. Geogr. Res.; 2012; 31, pp. 144-154.

10. Ettredge, M.; Gerdes, J.; Karuga, G. Using web-based search data to predict macroeconomic statistics. Commun. ACM; 2005; 48, pp. 87-92. [DOI: https://dx.doi.org/10.1145/1096000.1096010]

11. Cooper, C.P.; Mallon, K.P.; Leadbetter, S.; Pollack, L.A.; Peipins, L.A. Cancer Internet Search Activity on a Major Search Engine, United States 2001–2003. J. Med. Internet Res.; 2005; 7, e36. [DOI: https://dx.doi.org/10.2196/jmir.7.3.e36] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/15998627]

12. Xiang, Z.; Pan, B. Travel queries on cities in the United States: Implications for search engine marketing for tourist destinations. Tour. Manag.; 2011; 32, pp. 88-97. [DOI: https://dx.doi.org/10.1016/j.tourman.2009.12.004]

13. Park, S.B.; Ok, C.M.; Chae, B.K. Using Twitter Data for Cruise Tourism Marketing and Research. J. Travel Tour. Mark.; 2016; 33, pp. 885-898. [DOI: https://dx.doi.org/10.1080/10548408.2015.1071688]

14. Padhi, S.S.; Pati, R.K. Quantifying potential tourist behavior in choice of destination using Google trends. Tour. Manag. Perspect.; 2017; 24, pp. 34-47. [DOI: https://dx.doi.org/10.1016/j.tmp.2017.07.001]

15. Önder, I. Forecasting tourism demand with Google trends: Accuracy comparison of countries versus cities. J. Tour. Res.; 2017; 19, pp. 648-660. [DOI: https://dx.doi.org/10.1002/jtr.2137]

16. Siliverstovs, B.; Wochner, D.S. Google Trends and reality: Do the proportions match?. J. Econ. Behav. Organ.; 2018; 145, pp. 1-23. [DOI: https://dx.doi.org/10.1016/j.jebo.2017.10.011]

17. Zhang, Y.; Jin, X.; Wang, Y.; Liu, R.; Jing, Y. Characterizing Spatial-Temporal Variation of Cultural Tourism Internet Attention in Western Triangle Economic Zone, China. Land; 2022; 11, 2221. [DOI: https://dx.doi.org/10.3390/land11122221]

18. Zhang, X.; Chen, S.; Liu, X.; Wang, Q.; Li, Z. Spatial-Temporal Characteristics and Influencing Factors of Network Attention to Ancient City Destination: A Case of Pingyao. Econ. Geogr.; 2016; 36, pp. 196–202+207. [DOI: https://dx.doi.org/10.15957/j.cnki.jjdl.2016.07.026]

19. He, X.; Zhang, Y.; Liu, Y. Temporal and spatial characteristics of network attention to show cave: A case study of five beautiful show caves. Carsolog. Sin.; 2017; 36, pp. 275-282. [DOI: https://dx.doi.org/10.11932/karst20170216]

20. Wan, T.; Zhang, Z.; Li, S.; Liang, R. Research on the Temporal and Spatial Distribution of the Attention to Rural Tourism on Domestic Network. J. Southwest Univ.; 2022; 44, pp. 138-149. [DOI: https://dx.doi.org/10.13718/j.cnki.xdzk.2022.06.015]

21. Sun, Y.; Zhang, H.; Liu, P.; Zhang, J. Forecast of Tourism Flow Volume of Tourist Attraction Based on Degree of Tourist Attention of Travel Net-work: A Case Study of Baidu Index of Different Clients. Hum. Geogr.; 2017; 32, pp. 152-160. [DOI: https://dx.doi.org/10.13959/j.issn.1003-2398.2017.03.020]

22. Li, H.; Gao, H.; Song, H. Tourism forecasting with granular sentiment analysis. Ann. Tour. Res.; 2023; 103, 103667. [DOI: https://dx.doi.org/10.1016/j.annals.2023.103667]

23. Tang, H.; Xu, C. Spatio-temporal evolution and influencing factors of Chinese red tourism classic scenic spots network attention. J. Nat. Resour.; 2021; 36, pp. 1792-1810. [DOI: https://dx.doi.org/10.31497/zrzyxb.20210712]

24. Liang, L.; Fu, H.; Li, J.; Li, B. Spatio-temporal Dynamic Evolution and Influencing Factors of Net Celebrity City Network Attention: A Case of Xi’an. Sci. Geogr. Sin.; 2022; 42, pp. 1566-1576. [DOI: https://dx.doi.org/10.13249/j.cnki.sgs.2022.09.006]

25. Wang, C.; Lu, C.; Ba, D.; Ma, B.; Qin, Z. Spatio-temporal evolution and influencing factors of network attention of representative ski resorts in China. J. Nat. Resour.; 2022; 37, pp. 2367-2386. [DOI: https://dx.doi.org/10.31497/zrzyxb.20220912]

26. Zheng, E.; Xue, C.; Chen, G.; Zhang, Y.; Zou, J. Unveiling urban marathon development characteristics and urban growth strategies in China: Insights from time series analysis of Baidu Search Index. PLoS ONE; 2023; 18, e0287760. [DOI: https://dx.doi.org/10.1371/journal.pone.0287760]

27. Gao, N.; Zhang, X.; Wang, Y. Spatio-temporal characteristics and influencing factors of Chinese red tourism network attention. J. Nat. Resour.; 2020; 35, pp. 1068-1089. [DOI: https://dx.doi.org/10.31497/zrzyxb.20200505]

28. Wood, S.A.; Guerry, A.D.; Silver, J.M.; Lacayo, M. Using social media to quantify nature-based tourism and recreation. Sci. Rep.; 2013; 3, 2976. [DOI: https://dx.doi.org/10.1038/srep02976]

29. García-Palomares, J.C.; Gutiérrez, J.; Mínguez, C. Identification of tourist hot spots based on social networks: A comparative analysis of European metropolises using photo-sharing services and GIS. Appl. Geogr.; 2015; 63, pp. 408-417. [DOI: https://dx.doi.org/10.1016/j.apgeog.2015.08.002]

30. Giglio, S.; Bertacchini, F.; Bilotta, E.; Pantano, P. Using social media to identify tourism attractiveness in six Italian cities. Tour. Manag.; 2019; 72, pp. 306-312. [DOI: https://dx.doi.org/10.1016/j.tourman.2018.12.007]

31. Paldino, S.; Bojic, I.; Sobolevsky, S.; Ratti, C.; González, M.C. Urban magnetism through the lens of geo-tagged photography. EPJ Data Sci.; 2015; 4, 5. [DOI: https://dx.doi.org/10.1140/epjds/s13688-015-0043-3]

32. Marine-Roig, E.; Anton Clavé, S. Tourism analytics with massive user-generated content: A case study of Barcelona. J. Destin. Mark. Manag.; 2015; 4, pp. 162-172. [DOI: https://dx.doi.org/10.1016/j.jdmm.2015.06.004]

33. Kim, K.; Park, O.; Yun, S.; Yun, H. What makes tourists feel negatively about tourism destinations? Application of hybrid text mining methodology to smart destination management. Technol. Forecast. Soc. Chang.; 2017; 123, pp. 362-369. [DOI: https://dx.doi.org/10.1016/j.techfore.2017.01.001]

34. Marine-Roig, E. Destination Image Analytics through Traveller-Generated Content. Sustainability; 2019; 11, 3392. [DOI: https://dx.doi.org/10.3390/su11123392]

35. Philander, K.; Zhong, Y. Twitter sentiment analysis: Capturing sentiment from integrated resort tweets. Int. J. Hosp. Manag.; 2016; 55, pp. 16-24. [DOI: https://dx.doi.org/10.1016/j.ijhm.2016.02.001]

36. Yang, L.; Li, D.; Du, L.; Han, X. Study on temporal and spatial variation of network attention in national park scenic spots: A case study of Pan-Changjiang River Delta. Resour. Dev. Mark.; 2017; 33, pp. 1142-1146. [DOI: https://dx.doi.org/10.3969/j.issn.1005-8141.2017.09.024]

37. Némethová, J.; Vilinová, K. Changes in the Structure of Crop Production in Slovakia after 2004 Using an Example of Selected Crops. Land; 2022; 11, 249. [DOI: https://dx.doi.org/10.3390/land11020249]

38. Tirpáková, A.; Vojteková, J.; Vojtek, M.; Vlkolinská, I. Using Fuzzy Logic to Analyze the Spatial Distribution of Pottery in Unstratified Archaeological Sites: The Case of the Pobedim Hillfort (Slovakia). Land; 2021; 10, 103. [DOI: https://dx.doi.org/10.3390/land10020103]

39. Wang, J.; Xu, C. Geodetector: Principle and prospective. Acta Geogr. Sin.; 2017; 72, pp. 116-134. [DOI: https://dx.doi.org/10.11821/dlxb201701010]

40. He, Q.; Zheng, X.; Xiao, X.; Luo, L.; Lin, H.; He, S. The Spatiotemporal Evolution and Influencing Factors of the Ceramics Industry in Jingdezhen in the Last 40 Years. Land; 2023; 12, 1554. [DOI: https://dx.doi.org/10.3390/land12081554]

41. Tang, R. A study of the effects and mechanisms of the digital economy on high-quality tourism development: Evidence from the Yangtze River Delta in China. Asia Pac. J. Tour. Res.; 2022; 27, pp. 1217-1232. [DOI: https://dx.doi.org/10.1080/10941665.2023.2174033]

42. Adamov, T.; Iancu, T.; Peț, E.; Popescu, G.; Șmuleac, L.; Feher, A.; Ciolac, R. Rural Tourism in Marginimea Sibiului Area—A Possibility of Capitalizing on Local Resources. Sustainability; 2022; 15, 241. [DOI: https://dx.doi.org/10.3390/su15010241]

43. Li, L.; Gao, Q. Researching Tourism Space in China’s Great Bay Area: Spatial Pattern, Driving Forces and Its Coupling with Economy and Population. Land; 2023; 12, 1878. [DOI: https://dx.doi.org/10.3390/land12101878]

44. Rong, H.; Tao, Z. Hotspot identification and cause analysis of rural tourism based on website data: Take Jiangsu province as an example. J. Nat. Resour.; 2020; 35, pp. 2848-2861. [DOI: https://dx.doi.org/10.31497/zrzyxb.20201203]

45. Qin, W.; Zhang, Y.; Li, S. Study on the spatio-temporal evolution of coastal city tourism of China. Geogr. Res.; 2014; 33, pp. 1956-1965.

46. Mu, X.; Guo, X.; Ming, Q.; Hu, C. Dynamic evolution characteristics and driving factors of tourism ecological security in the Yellow River Basin. Acta Geogr. Sin.; 2022; 77, pp. 714-735. [DOI: https://dx.doi.org/10.11821/dlxb202203015]

47. Kong, X.; Fu, M.; Jiang, P. Spatial pattern and optimization zoning of characteristic villages based on tourism space in China. Acta Geogr. Sin.; 2023; 78, pp. 2554-2573. [DOI: https://dx.doi.org/10.11821/dlxb202310011]

48. Ivankova, V.; Gavurova, B.; Bačík, R.; Rigelský, M. Relationships between road transport infrastructure and tourism spending: A development approach in European OECD countries. Entrep. Sustain. Issues; 2021; 9, pp. 535-551. [DOI: https://dx.doi.org/10.9770/jesi.2021.9.2(35)]

49. Huang, Z.; Zhang, Y.; Jia, W.; Hong, X.; Yu, R. The research process and trend of development in the New Era of rural tourism in China. J. Nat. Resour.; 2021; 36, pp. 2615-2633. [DOI: https://dx.doi.org/10.31497/zrzyxb.20211012]

50. Hoffmann, F.J.; Braesemann, F.; Teubner, T. Measuring sustainable tourism with online platform data. EPJ Data Sci.; 2022; 11, 41. [DOI: https://dx.doi.org/10.1140/epjds/s13688-022-00354-6]

51. Li, S.; Li, S.; Huang, Z.; Wang, M.; Teng, L. Spatial differentiation characteristics and cause analysis of vitality intensity of China’s 5A-level scenic spots based on Tencent’s location big data. Sci. Geogr. Sin.; 2023; 43, pp. 1239-1248. [DOI: https://dx.doi.org/10.13249/j.cnki.sgs.2023.07.011]

52. Solazzo, G.; Maruccia, Y.; Ndou, V.; Del Vecchio, P. How to exploit Big Social Data in the COVID-19 pandemic: The case of the Italian tourism industry. Serv. Bus.; 2022; 16, pp. 417-443. [DOI: https://dx.doi.org/10.1007/s11628-022-00487-8]

53. Pramana, S.; Paramartha, D.Y.; Ermawan, G.Y.; Deli, N.F.; Srimulyani, W. Impact of COVID-19 pandemicon tourism in Indonesia. Curr. Issues Tour.; 2022; 25, pp. 2422-2442. [DOI: https://dx.doi.org/10.1080/13683500.2021.1968803]

54. Yang, J.; Yang, R.; Chen, M.-H.; Su, C.-H.; Zhi, Y.; Xi, J. Effects of rural revitalization on rural tourism. J. Hosp. Tour. Manag.; 2021; 47, pp. 35-45. [DOI: https://dx.doi.org/10.1016/j.jhtm.2021.02.008]