Abstract
As one of the three pillar industries of tourism, the hotel industry has developed rapidly in recent years, especially the construction and development of urban star hotels. The major star hotel brands have poured in one after another, and a large amount of funds have been invested in the star hotel market. The continuous construction of various types of star hotels has formed the geographical agglomeration phenomenon of star hotels in Chinese cities. This phenomenon has not only brought the advantages of regional economies of scale and brand building, but also brought the competition between star hotels. Under the guidance of market demand, tourism can be created through industrial integration and functional combination to develop the tourism industry, which makes the spatial distribution of tourism hotels closely related to transportation, consumer demand, urban environment, and even industrial policy. In recent years, artificial intelligence methods are more and more applied to tourism, mainly including rough set method, genetic algorithm, fuzzy time series, grey theory, artificial neural network, and support vector machine. Based on artificial intelligence technology, this paper studies the spatial distribution characteristics and configuration of tourism hotels. The distance of the peak value of tourism hotel spatial distribution characteristics is 4.8 km and 6.6 km, respectively, and the distance of the peak value of natural tourism resources is 0.98 km. Because the spatial distribution characteristics of cultural, special, and tourism hotels are restricted by their own quantity, resource endowment, and inherent attributes, the spatial scope of location layout is relatively small. The spatial distribution characteristics and configuration of tourist hotels can also timely monitor and prevent various emergencies, prevent tourism safety accidents, and improve the ability of tourism emergency management by using artificial intelligence and other technologies. With the continuous optimization of artificial intelligence technology and the deepening of cooperation between tourism disciplines and interdisciplinary, the research on tourism big data driven by research problems will be more mature.
1. Introduction
With the increase of people’s leisure time, the pursuit of health, and the rise of leisure and holiday tourism, the traditional sightseeing tourism can no longer meet the needs of modern people. The trend of tourism consumption is changing from sightseeing tourism to leisure and holiday tourism, and leisure tourism is gradually becoming a consumption hotspot [1, 2]. As one of the three pillar industries of tourism, the hotel industry has developed rapidly in recent years, especially the construction and development of star-rated hotels in cities. With the influx of major star-rated hotel brands, a large amount of money has been invested in the star-rated hotel market, and various types of star-rated hotels have been continuously built, which has formed the phenomenon of geographical agglomeration of star-rated hotels in Chinese cities. This phenomenon has brought advantages such as regional economies of scale and brand building, as well as competition among star-rated hotels [3]. Traditional tourism resources mostly exist in the form of natural or cultural heritage, and their spatial layout is fixed. However, leisure tourism resources are moldable and innovative. Under the guidance of market demand, leisure tourism resources can be created through industrial integration and functional compounding to develop leisure tourism industry, which makes the spatial distribution of leisure tourism resources closely related to transportation, consumer demand, urban environment, and even industrial policies [4, 5]. From the related research point of view, at present, most of the research on tourist hotels and restaurants starts from the micro-level, and the research from the spatial perspective is less. Only a few scholars have analyzed the scale of tourist hotels, the rationality of spatial structure, and the supply characteristics, and the research on the spatial configuration of domestic tourist hotels and hotels is relatively weak [6]. In this paper, five-star hotels in high-grade tourist hotels are taken as the research object, and combined with the development of inbound tourism, the spatial distribution characteristics, regional supply, and allocation principles are preliminarily discussed.
In recent years, artificial intelligence methods are more and more applied to tourism, mainly including rough set method, genetic algorithm, fuzzy time series, grey theory, artificial neural network, and support vector machine. The term artificial intelligence was born in the Summer Seminar on artificial intelligence held at Dartmouth College in 1956. So far, it has a development history of more than 60 years [7, 8]. Since 2006, with the development of big data, high-performance parallel computing, deep learning, and other technologies, data has accumulated rapidly, computing power has been greatly improved, algorithm models have continued to evolve, and artificial intelligence has ushered in a new development climax. Artificial intelligence is becoming an important field related to national economy, social development, and national security [9, 10]. The biggest advantage of artificial intelligence is that it has no strict requirements for additional information such as probability distribution of data and has better inclusiveness and adaptability. An objective fact closely related to the development of China’s tourism industry is that China’s Internet and its applications have made remarkable achievements in the world and become a “booster” to improve the quality of China’s economic development and promote supply side reform. Because of artificial intelligence technology, the traditional mathematical-statistical model has strict requirements on assumptions and sample data categories when the model is applied to practice, and only using mathematical formulas and theories to describe enterprise business performance can not comprehensively and effectively explain the complex and changeable situation faced by enterprises in the real society, resulting in the reduction of the accuracy and reliability of model classification and evaluation [11, 12].
Application of artificial intelligence in hotel scene, tourism distribution, outbound travel and travel itinerary planning are as follows: The relationship between “technology and people”, such as the influence of human-computer interaction on service, human-computer mixed organization; the relationship between technology and emotion in service; the destruction of service value caused by the conflict between technical systems; and the interaction between customers and technology [13, 14]. At the same time, with the continuous optimization of artificial intelligence technology and the deepening of cooperation between tourism disciplines and inter-disciplines, the research on tourism big data driven by research problems will become more mature. The influence of artificial intelligence on tourism enterprises, the relationship between village houses and local villagers, the management of the new generation of hotel employees and the aging employment, etc. At the same time, the difficult problems in practice have not been paid enough attention in the existing domestic academic research, such as the stigmatization of tour guides, the zero-negative tour fee of travel agencies, the turnover of hotel employees, and the service quality. The spatial distribution characteristics and configuration of hotels can also monitor and prevent various emergencies in time, prevent tourism safety accidents, and use artificial intelligence and other technologies to improve tourism emergency management ability [15].
This paper studies and innovates the above problems from the following aspects: (1)This paper puts forward the spatial distribution characteristics and configuration model of tourist hotels based on artificial intelligence technology. With its parallel processing ability, self-learning, self-organization, self-adaptive ability, and good fault tolerance, artificial intelligence network model adapts to many problems such as incomplete tourism data information, many influencing factors, large uncertainty, and nonlinearity and makes up for the shortcomings of traditional prediction methods(2)The spatial distribution characteristics and configuration system of tourist hotels based on artificial intelligence technology are constructed. Making rational use of artificial intelligence system to plan tourism routes based on these data is a hot spot in recent research work. The advantage of this kind of work is that it can quickly get the feasible solution in line with the actual situation and help users make travel planning, but the difficulty is to make rational use of the multisource data of artificial intelligence technology to accurately mine the historical behavior trajectory of users. As a nonparametric statistical method, kernel density estimation can directly reflect the agglomeration degree of the research object by estimating the occurrence probability of point elements in different geographical spaces
The paper is divided into five parts, and the organizational structure is as follows.
The first chapter introduces the research background and current situation of the spatial distribution characteristics and allocation of tourism hotels and puts forward and summarizes the main tasks of this paper. The second chapter introduces the spatial distribution characteristics and allocation of tourist hotels. The third chapter introduces the algorithm and model of artificial intelligence technology. The fourth chapter introduces the spatial distribution characteristics of tourist hotels and the implementation of the configuration system and compares the performance of the system through experiments. The fifth chapter is the summary of the full text.
2. Related Work
2.1. Research Status at Home and Abroad
Bhati et al. Proposed that as a high-end tourist hotel, the service quality, housing environment, and health status of five-star hotels have always been highly affirmed by domestic and foreign tourists. Among all kinds of domestic foreign-related tourist hotels, the economic efficiency is high, the operating efficiency is good, the operating advantages are the most prominent, and the room rental rate remains above 65% all year round, which is the type with the highest room rental rate among all star hotels. Therefore, the spatial distribution characteristics of five-star hotels can fully reflect the development level and reception strength of local tourism reception industry [16]. Leh et al. proposed that from the perspective of spatial distribution types, the spatial distribution of geographical things mainly includes four types: point distribution type, linear distribution type, discrete regional distribution type, and continuous regional distribution. The latter two spatial distribution types can be attributed to area distribution. Therefore, the spatial distribution types of all geographical things can be attributed to three types: point, line, and area [17]. Karl puts forward the definition of tourism resources. All kinds of things and factors that can attract people’s activities, can be developed and utilized for tourism, and can produce economic, social, and environmental benefits are the theoretical basis of this paper [18]. Rai et al. proposed that the rapid development of the spatial distribution characteristics of tourist hotels and the consumption demand of tourism configuration for information make tourism informatization and information application the trend and inevitable choice of the development of tourism industry [19]. Li et al. put forward that the research on tourism resources mainly has the following characteristics: The research content mainly focuses on the exploration of tourism resources development mode and the evaluation of tourism resources development status and tourists’ behavior, and there is little research on the spatial distribution and formation mechanism of tourism resources; in terms of research methods, most of them use qualitative methods, while quantitative analysis and data visualization methods are less involved; as for the research object, it mainly focuses on single types and single elements such as ancient towns, urban parks, and characteristic blocks, while the types of tourism resources are diverse. Only taking a single element as the research object cannot comprehensively understand the regional layout characteristics of tourism resources; in terms of research scale, most of them start from the perspective of macro scale, and the research on the meso- and microscale of the spatial distribution of tourism resources is still insufficient [20]. Jiang proposed that there are various charging standards in the tourism platform, which can reasonably guide tourists to arrange their own tourism routes according to their own economic situation, and the platform also provides online payment methods to provide tourists with different tourism service experiences [21]. Meidute-Kavaliauskiene et al. put forward that the five-star hotels are closely related to inbound tourism, which can be abstracted as point elements in space. The spatial distribution characteristics of point elements are equal, random, and aggregated. Here, the concentration index is used to measure the spatial distribution characteristics of five-star hotels in China [22]. Majumdar et al. put forward that the tourism industry provides an automated service platform, so its platform not only includes the spatial distribution characteristics and configuration of tourist hotels, but also includes the routes and expenses of major tourist attractions [23]. Varakantham et al. put forward that the spatial distribution characteristics and configuration of tourist hotels are the practical innovation and integrated innovation of information technology in tourism and the systematic and intensive management reform to meet the personalized service experience needs of tourists and provide high-quality and high-satisfaction services and to realize the integrated sharing and effective utilization of tourism resources, social resources, and environment in a coordinated way [24]. Hu et al. put forward that through the reform of the spatial distribution characteristics and configuration of tourist hotels, the tourism industry can actively or passively break the original development model and explore the path of developing smart tourism. The application and innovation at all levels of cities also provide a supporting environment and practical experience for tourism [25].
2.2. Research Status of Spatial Distribution Characteristics of Artificial Intelligence
Based on artificial intelligence technology, this paper studies the spatial distribution characteristics and configuration of tourist hotels that analyze the spatial distribution characteristics of tourist hotels and further analyze the influencing factors leading to their spatial differentiation by using geographic detectors to provide a reference for the development of tourism hotel business areas and areas to be improved; realize the rational allocation between tourism hotels and various resources; promote the transformation, upgrading, and optimal layout of the tourism hotel industry; and promote the high-quality development of regional rural tourism. On the one hand, the research on its spatial distribution and change characteristics can reflect the change trend of China’s high-end hotel supply; on the other hand, it also reflects the gap in the development of inbound tourism in various regions. Through the development of inbound tourism under artificial intelligence technology, the development speed of domestic five-star hotels is accelerated. The spatial homogenization development trend of inbound tourism guides and coordinates the spatial distribution of five-star hotels. The change of space supply of China’s five-star hotels is highly consistent with the development trend of the spatial pattern of inbound tourism. Promoting the spatial distribution characteristics and allocation of tourist hotels through artificial intelligence technology and advanced development concepts can improve the quality of tourism services; optimize tourism management; help to save tourists, enterprises, and governments’ tourism time, space, and information transmission costs; reduce resource waste; and comprehensively realize and improve the social, economic, and environmental benefits of the tourism industry. In addition, tourists can also evaluate the spatial distribution characteristics and configuration services of tourist hotels through artificial intelligence technology so that the tourism bureau can make reasonable improvement through customer feedback, so as to improve the management level of the tourism industry and improve tourism services, so as to meet the needs of tourists.
3. Algorithms and Models of Artificial Intelligence Technology
The traditional mathematical statistical model is relatively mature in the field of classified evaluation of business performance, forming a relatively complete research system. However, due to the strict requirements of assumptions and the limitation of data types and quantities, when the traditional mathematical statistical model is applied to practical problems, the accuracy and reliability of model classification and evaluation will be reduced. By adjusting the relationship between a large number of internal nodes, the purpose of information processing can be achieved. Artificial neural network has a parallel distributed information processing structure and the ability of self-learning and self-adaptation. By providing a batch of mutually corresponding input and output data in advance, it can analyze and master the potential law between them and accumulate information. Artificial intelligence algorithm can continuously learn and train a large number of data of different types and dimensions and accurately and reasonably reflect different complex environments. At present, the artificial intelligence algorithm model has successfully attracted the attention of a large number of researchers and is widely used in sales forecasting, stock analysis, customer service, and other fields. This algorithm can quickly learn and train from a large number of data of different types and dimensions and get results that are more suitable for complex environment, which has attracted great attention of scholars in various fields. The research on classified evaluation of enterprise performance has also begun to change from traditional mathematical statistics method to artificial intelligence algorithm. When solving the problem of spatial distribution characteristics and configuration of tourist hotels, in addition to building detailed user portraits and mining users’ behavior habits and related preferences, it is also necessary to consider users’ contextual information and related tourism information in order to further improve the accuracy of recommendation results. Based on this understanding, the spatial distribution characteristics of tourist hotels and the overall framework of the configuration system are analyzed, as shown in Figure 1.
Combined with artificial intelligence technology, this paper uses four algorithms and models: logistic regression, support vector machine, random forest tree, and gradient lifting tree to compare and analyze the effects of the four algorithms and models in classifying the spatial distribution characteristics and configuration of tourism hotels in China. Although logistic regression is a traditional mathematical statistical model, its dependent variable type is expressed in binary form. Therefore, when the logistic regression model classifies the sample data, it determines the best parameters through observation, learning, and continuous training of the sample data, so as to effectively classify and predict the samples. The overall model design is shown in Figure 2.
The nearest neighbor hierarchical cluster analysis method is an analysis method based on the distance of points to explore the hot spots of the spatial distribution of point data. It is mainly used to describe the overall distribution situation of point data. In practice, the nearest neighbor index is generally used to judge whether the spatial distribution of point data belongs to agglomeration type, and then the nearest neighbor hierarchical cluster analysis method is used to explore the hot spot agglomeration area. The calculation formula can be expressed as where is the nearest neighbor distance ; is the expected average nearest neighbor distance ; is the number of sample points (pieces); and is the study area . means that the sample points are clustered, and the smaller the value, the more clustered; indicates that the sample points are distributed uniformly and discretely; indicates that the sample points are randomly distributed.
The nearest neighbor index can only describe the overall distribution pattern of point data, and cannot judge the distribution characteristics of various leisure tourism resources on different spatial scales, while function can analyze the distribution pattern of spatial point elements on different spatial scales. This paper uses function to analyze the spatial agglomeration mode of various tourism resources in Chengdu on different spatial scales. The calculation formula can be expressed as where represents the agglomeration degree of tourist resources and function is constructed to keep the variance stable; is the number of various leisure tourism resources (individual); is the distance between a certain type of tourism resource point and resource point within the distance ; the relationship between and can test the spatial distribution pattern of various leisure tourism resources within the range of distance . It means that the leisure tourism resources are clustered, means that the leisure tourism resources are randomly distributed, and means that the leisure tourism resources are dispersed.
In this paper, the influencing factors of spatial distribution pattern of tourism resource density are discussed with the help of geographical detectors, and the calculation formula can be expressed as follows: where is the influence factor has on the density of leisure tourism resources, and the larger the values of and , the greater the influence factor has on the density of tourism resources; is the stratification of tourism resource density or influencing factor ; and are the number of units (units) and variance of layer , respectively; and and are the unit number and variance of the whole study area, respectively.
Further, according to the research of and , combined with the characteristics of tourism, this paper constructs the following overall output model of regional tourism enterprises:
It is assumed that the distribution of tourism enterprises in each region is uniform and the total output of tourism enterprises per unit area of the region is a function of the tourism factor input per unit area, the total scale of regional tourism, and the relevant technical effect . Then, according to the production function, this paper will rewrite it into the following model:
where is the total labor force and capital input of tourism enterprises in region, respectively, and and represent the output elasticity coefficient of tourism enterprise income to labor force and capital input per unit area, respectively. measures the impact of the Internet and related applications on the output of tourism enterprises.
Tourism intensity is defined as the level of tourism economic activity per unit area. measures the impact of tourism intensity on the income of tourism enterprises per unit area. represents the response coefficient of tourism enterprise income per unit land area to the overall tourism intensity. Further, the total income level of regional tourism enterprises can be expressed as the income per unit area multiplied by the total area of the region, i.e.,
where is the total income level of tourism enterprises in regional ; then, dividing both sides by the number of labor forces can obtain the labor productivity of regional tourism enterprises: where represents the labor productivity of tourism enterprises, which reflects the income capacity and level of unit tourism labor force. The equation shows that the existence of the scale effect of tourism development means that the labor productivity of regional tourism enterprises is not only a function of its general factor input, but also a function of the diversified development and industrial intensity of tourism in the Internet application and its region.
Type represents the labor productivity of tourism enterprises, which reflects the income capacity and level of unit tourism labor force. The equation shows that the existence of the scale effect of tourism development means that the labor productivity of regional tourism enterprises is not only a function of their general factor input, but also a function of the application of Internet and the diversified development and industrial intensity of tourism in their region:
is time; , , and , respectively, represent the diversified development level of Internet and tourism and their interaction items; and and represent unit labor capital and tourism intensity, respectively. is the coefficient to be estimated.
In the artificial intelligence algorithm, the digital codes of several solved problems, namely, chromosomes, are generated randomly to form the initial population; Through the fitness function, each individual is given a numerical evaluation, the individuals with low fitness are eliminated, and the individuals with high fitness are selected to participate in the genetic operation. After the genetic operation, the collection of individuals will form a new population of the next generation, and the new population will undergo the next round of evolution. With its parallel processing ability, self-learning, self-organization, self-adaptive ability, and good fault tolerance, the artificial intelligence network model adapts to many problems such as incomplete information of tourism data, many influencing factors, great uncertainty, and nonlinearity and makes up for the shortcomings of traditional forecasting methods.
4. Realization of Spatial Distribution Characteristics and Configuration of Tourist Hotels
4.1. Spatial Distribution Characteristics and Configuration System Based on Artificial Intelligence Technology
The biggest factor for the spatial distribution characteristics and configuration system of tourism under special artificial intelligence technology is street vitality, followed by the density of tourists, the distance from the city center, and the density of local residents. A large proportion of special resources are business clubs, which essentially provide a platform for business information exchange. The high street vitality is conducive to information exchange, while the areas close to the city center not only have high street vitality, but also have relatively concentrated tourists and local residents. Therefore, these factors are similar to special tourism resources in spatial distribution. Scientific spatial distribution characteristics and configuration system planning can not only help travelers to make their own tour routes according to their own time and budget, but also enhance their travel experience. Artificial intelligence technology enables travelers to spend more time and energy on the tour process. In the single route solution of the tourism route planning problem, the artificial intelligence technology can be used to model the spatial distribution characteristics of tourist hotels and the allocation of the single-objective traveling salesman problem, which can increase the revenue objective, and link the connection between nodes with revenue and travel cost. The goal is to find a circuit on a subset of all nodes, which can maximize the profit and minimize the travel cost. By further analyzing the spatial distribution characteristics and configuration system of tourist hotels with artificial intelligence technology, it is found that resource endowment and traffic convenience have little influence on them, and they have not passed the significance test, which shows that they have little correlation with the distribution of special tourism resources. However, this kind of work cannot completely model the various factors of real-life tourism route planning. On the one hand, because tourism is a dynamic process, there are many uncertain factors in this process; On the other hand, when the geographical scope of the point of interest is large, the point of interest can no longer be modeled as a point, such as a sightseeing river, and the starting point and the ending point of the point of interest may be far apart. These data share a large number of spatial distribution features with users, and data such as travel experience and travel photos in the configuration system together form tourism big data.
Making rational use of artificial intelligence technology to plan tourism routes based on these data is a hot spot in recent research work. The advantage of this kind of work is that it can quickly get the feasible solution in line with the actual situation and help users make travel planning, but the difficulty is to make rational use of the multisource data of artificial intelligence technology to accurately mine the historical behavior trajectory of users. The nearest neighbor index analysis only describes the spatial distribution of point elements from the perspective of mathematical statistics, which cannot truly reflect its spatial distribution pattern. As a nonparametric statistical method, kernel density estimation can directly reflect the agglomeration degree of the research object by estimating the occurrence probability of point elements in different geographical spaces. The degree of agglomeration of points is related to the occurrence probability of the research object. The higher the occurrence probability, the more agglomeration. Among the factors affecting the spatial distribution characteristics and allocation system of natural tourism resources, except the tourist factor, all factors failed to pass the significance test. The possible reasons are as follows: on the one hand, natural tourism resources are spatially immovable, and their spatial distribution is less affected by social and economic factors; on the other hand, because the study area of this paper is the main urban area, does not include all natural tourism spots, and the sample size is small, there are differences in the results.
4.2. Experimental Results and Analysis
In this experiment, according to the predecessors’ classification of tourist hotels, combined with the actual situation of tourist hotels, the data was cleaned according to the types of POI data and finally reclassified according to three-star hotels, four-star hotels, and five-star hotels, as shown in Table 1.
It can be seen from Table 1 that the scale of three-star hotels is the smallest compared with four-star hotels and five-star hotels, so it is abstracted as a point for analysis. In addition, because this paper takes the tourist hotel point data as the research object and the resource monomer within the scope of this kind of resources as the research object, the proportion of five-star hotels is the highest among the three categories, which verifies that the agglomeration characteristics of five-star hotels are obvious in tourist hotels. This experiment uses the nearest neighbor index to analyze the agglomeration of different types of tourist hotels, as shown in Table 2.
It can be seen from Table 2 that there is little difference in spatial agglomeration of tourist hotels. The nearest neighbor index of tourist hotels as a whole and all kinds of tourist hotels are less than 1, and the test values are all less than -2.58, which is a typical agglomeration model after passing the test at 1% significance level. Judging from the agglomeration degree of tourist hotels, it is five-star hotels > four-star hotels > three-star hotels. Because of the large number and concentrated distribution of five-star hotels and general tourist hotels, the agglomeration degree is higher, while the three-star hotels have less tourism volume and are far apart from each other, so the agglomeration degree is weaker than other types of tourism resources.
In this experiment, Ripley’s K function is used to discriminate and configure the spatial distribution of various types of tourist hotels in multiscale. Three comparisons are made in this experiment, and the experimental results are shown in Figures 3, 4, and 5.
From Figures 3–5, it can be seen that tourist hotels all obey the agglomeration distribution on different spatial scales and there is significant agglomeration. This paper verifies the analysis results of the previous nearest neighbor index from another angle. From the changing trend of L(d) function curve, the spatial distribution characteristics of all kinds of tourist hotels are similar, and there is no phenomenon of no peak and double peak. From the distance of the peak, the spatial distribution characteristics of the total tourist hotels reach the aggregation peak at 30.5 km, which is higher than that of any single type of tourist hotels, indicating that the overall aggregation intensity of the spatial distribution characteristics of tourist hotels is enhanced under the joint action of the spatial distribution characteristics of tourist hotels. The spatial scale of different types of tourist hotels is obviously different. The distance from the peak value of tourism resources for recreation and entertainment is the largest, which is 11.7 km, which shows that tourism resources for recreation and entertainment show agglomeration characteristics in a large spatial scale and the spatial range of location layout is large. The spatial distribution features of tourist hotels are 4.8 km and 6.6 km away from the peak value of natural tourism resources and 0.98 km away from the peak value of natural tourism resources. Because the spatial distribution features of cultural, special, and tourist hotels are restricted by their own quantity, resource endowment, and inherent attributes, the spatial scope of location layout is relatively small.
In order to further reflect the change trend and characteristics of urban tourism development efficiency, this experiment introduces the coefficient of variation to analyze the urban tourism development efficiency from 2016 to 2022. The experimental results are shown in Figures 6 and 7.
From Figure 6–7, it is not difficult to find that from 2016 to 2022, on the whole, the variation index values of comprehensive efficiency, pure technical efficiency, and scale efficiency of urban tourism development show fluctuating characteristics, which shows the instability of urban tourism development process. Among them, the variation index of comprehensive efficiency and scale efficiency fluctuates significantly. The change trend of pure technical efficiency variation index shows that there are differences in the investment and introduction of advanced technology among cities in the region and the investment proportion of tourism related industries in the field of technology.
5. Conclusions
In recent years, the research on the spatial distribution of domestic star hotels has gradually increased and achieved some results. The agglomeration effect produced by hotel spatial agglomeration, as an important influencing factor of spatial distribution characteristics or an important content of regional tourism agglomeration, appears more frequently, and scholars have less direct research on hotel spatial agglomeration. According to the influencing factors of the spatial distribution characteristics and configuration of tourist hotels based on artificial intelligence technology, the traffic convenience ranks first. Therefore, we should strengthen the construction of traffic facilities and urban traffic management, improve the accessibility of the spatial distribution characteristics and configuration of tourist hotels, and reduce the travel time cost of tourists; secondly, spatial agglomeration is also an important factor affecting the spatial distribution characteristics and allocation differences of tourism hotels. Therefore, in areas with weak tourism industry development, relevant preferential policies can be given through government intervention to attract the inflow of spatial distribution characteristics and allocation of tourism hotels, so as to improve the agglomeration degree of spatial distribution characteristics and allocation of tourism hotels. The distance of the peak concentration of tourism hotel spatial distribution characteristics is 4.8 km and 6.6 km, respectively, and the distance of the peak concentration of natural tourism resources is 0.98 km. Because the spatial distribution characteristics of cultural, special, and tourism hotels are restricted by their own number, resource endowment and inherent attributes, the spatial scope of location layout is relatively small. In the follow-up research of artificial intelligence technology, we will further enrich the data sources, expand the scope of the research area, increase the data time span, and deeply explore the spatial distribution characteristics and allocation spatial pattern, temporal and spatial evolution, and the causal relationship behind the tourism hotels in combination with the “small data” of traditional surveys and interviews, so as to better guide the development of urban tourism industry. These data and a large number of spatial distribution characteristics shared by users and data such as tourism experience and travel photos in the configuration system jointly form tourism big data.
Data Availability
The data used to support the findings of this study are included within the article.
Conflicts of Interest
No competing interests exist concerning this study.
Acknowledgments
The authors would like to show sincere thanks to those techniques who have contributed to this research.