Delineating Landscape Features Perception in Tourism-Based Traditional Villages: A Case Study of Xijiang Thousand Households Miao Village, Guizhou

.


Introduction
Since 2012, the Ministry of Housing and Urban-Rural Development in China, alongside various other governmental bodies, has unveiled six distinct batches of traditional villages, culminating in the designation of 8155 villages [1].This extensive array and the diverse nature of traditional villages represent a cornerstone of the Chinese nation's cultural heritage, esteemed as one of its most precious assets.Traditional village landscapes have long been the focus of both scholarly inquiry and practical interest, reflecting a deepseated appreciation for their historical and cultural significance [2][3][4][5].These landscapes are distinguished from the notions of "rural landscape," "vernacular landscape," and "rural scenery," embodying a regional testament to the historical continuum of human activities and beliefs within the framework of a traditional agricultural civilisation.The essence of traditional village landscapes, encapsulated as regional "historical witnesses to human endeavours and convictions amidst the backdrop of traditional agricultural society," is sculpted through grassroots dynamics.These landscapes are imbued with attributes that are historical, invaluable, and distinctly emblematic of both national identity and rural ethos.Their formation is a testament to a bottom-up confluence of factors, thereby highlighting their profound historical, valuable, national, and rural characteristics [6][7][8].With the introduction of the rural revitalisation strategy, rural tourism has therefore flourished, and traditional villages have become popular tourist destinations due to their rich cultural landscape resources and natural landscape scenery, providing opportunities for the sustainable development and transformation of traditional villages [9,10].Meanwhile, sustainable tourism development is pivotal for the revitalisation of traditional villages, serving as a key component in their ongoing transformation and growth.
Previous studies on traditional village landscapes have mostly relied on field surveys [11] and in-depth interviews [12] to obtain data.However, field surveys often only reflect the landscape perception of the investigator or researcher, and in-depth interviews can only collect limited opinions from local residents, with a long period of time and serious regional restrictions, so it is difficult to obtain a large amount of data by traditional methods in the study of traditional village landscapes for tourism.Although the rapid development and application of big data and the internet of things have spawned many researchers and scholars to collect and evaluate landscape features using computational technologies such as GIS [13], these studies have little involvement of experienced users and do not incorporate a large number of tourists' online evaluation texts into data use, and researchers have not solved the problem of how to accurately and efficiently obtain a large number of tourists' comments, which makes it difficult to confirm the feasibility of utilising online comment data in the study of traditional village landscape features.In order to describe landscape features in a comprehensive and systematic way, it is indispensable for users to participate in the whole process and evaluate their perception after experience, which is also emphasised by the European Landscape Convention [14].
Therefore, the aim of this study is to improve the methodology for traditional village landscape features research using tourists' online comment data, taking Xijiang Thousand Households Miao Village as an example, to grasp the public's current main types of activity participation and the emotional tendencies generated in the process of traditional village tourism experience, and at the same time, to sort out the characteristics of Xijiang Thousand Households Miao Village landscape features under the tourists' perspectives, and to re-evaluate the value of the landscape resources of the traditional villages in a bottom-up manner.This method will also spotlight the challenges faced by managers in the landscape planning, protection, and utilisation processes, offering strategic recommendations for improvements.On the one hand, the study convincingly proves that online textual data can serve as an effective source of information for the preservation and management of village landscapes.On the other hand, it promotes the in-depth study of tourismbased traditional village landscapes in ethnic minority areas.In view of the phenomenon of remodelling the characteristics of village landscapes and the gradual loss of the uniqueness of village landscapes in the development of tourism, it provides insights into the challenges encountered by traditional village landscapes in changing times and then strengthens the management system of village landscapes, which is of great practical significance for the research on the conservation and development of other tourism-based traditional village landscapes in China.

Studies Related to the Landscape Features Perception
Landscape character is the identifiable, distinctive, and consistent elements of a landscape, not used to describe two landscapes as better or worse but to emphasise that one landscape is different from another [15,16].The study of landscape features has a long history, and experts and scholars in various fields around the world have successively invested in the study of landscape features [17][18][19][20][21].In the UK, Landscape Character Assessment (LCA) [22] has been an important tool for rural landscape conservation since the 1990s.Carys Swanwick and Christine Tudor provided a clear definition of landscape character in an official document issued by Natural England, an approach that has contributed significantly to the conservation of European landscapes [15].Van Eetvelde and Antrop obtained two different scales of Belgian landscape character types by creating a GIS database with a cross-regional and transboundary perspective [23].In Australia, Brown combines the physical elements and cultural features of the landscape with potential land uses to assess the landscape's scenic resource value, visual value, and cultural value in order to conserve Australia's regional landscape character [24].In China, the earliest research on landscape features can be traced back to 1998, when scholars such as Shi Gaojun worked on the collection and development of landscape features on the tourism resources of the famous mountains in Jiangsu Province, China, laying the foundation for the sustainable development and perpetual utilisation of their subsequent landscape tourism [25].In 2001, the City Planning Department of Hong Kong, China, carried out a complete landscape character mapping and assessment exercise for the whole of Hong Kong and produced a landscape character map [26].Zhu Yiting et al. evaluated the diversity of landscape types and landscape characters of the Cocotocai World Geopark using the Scenic Beauty Evaluation (SBE) method and GIS analysis technique [27].At present, scholars have been investing in the evaluation of rural landscape characters, exploring the way of collecting and evaluating landscape characters in China, so as to carry out landscape protection and management work in a more reasonable way [28,29].Landscape characters have multiple values, and landscape information can only be observed, received, and then perceived in the brain by a human being in order to make a description that is consistent with landscape characters [30].However, a commonality can be summarised from the known landscape character studies; even though professional landscape research organisations are using the LCA [31][32][33] methodology, they often tend to choose to start with aspects of regional culture and natural environment for the study of landscape features in different regions.They seldom involve the participation of experiential users in the process of landscape character collection.
The study of landscape feature perception can be understood as the study, categorisation, and even assessment of landscape features through the experiences of different groups of people who experience them [34].Since the 18th century, the concept of "landscape" has been adopted by garden designers."Landscape" refers to their design of the overall scene composed of buildings and natural environment.The understanding of the landscape during this period was based on visual aesthetics perception, with the main emphasis on comprehensive intuitive visual perception [35].In 2000, the European Landscape Convention stated that a landscape is a perceived area whose characteristics are the result of human and natural factors and their interactions [36].In his research on rural tourism strategies, Lin Yuanjun argues that landscape perception is a comprehensive response to tourists' impressions, feelings, and evaluations of the landscape environment [37].It is worth mentioning that this perception is not only on the visual level.Zhan Kaiyuan found in his study of landscape perception in urban parks that people's landscape perception, in addition to visual perception, mainly includes two aspects of soundscape perception and aroma perception [38].From these studies, we can find that landscape feature perception studies are centred on the elements that can be perceived and that people are the only initiators of feedback perception.The involvement of a large number of tourists is a primary prerequisite for the description of the delineating landscape perception feature and requires visitors to complete their experiences in the landscape environment and give authentic feedback to ensure the authenticity of landscape perceptions.

Studies Related to Traditional Village Landscapes
Currently, research on traditional village landscapes has shifted from focusing on a certain landscape element (e.g., terrain [39][40][41], forests [2,42], rivers [43], etc.) to a more comprehensive landscape type (e.g., the mechanism of settlement generation [44,45], building wisdom [46,47], activity behaviour [48], etc.).The focus of research has gradually shifted to the inheritance and protection of village features and cultural genes [49,50], and the exploration of village tourism models [10].In the field of village conservation and utilisation, Petrevska used an exploratory sequential design to analyse the possibilities of using traditional architecture to promote rural prosperity and stimulate economic development through tourism, and she proposed an innovative approach to revitalising villages through traditional architecture [9,51].Fu and other scholars used GIS and RS technologies to construct a spatial database of traditional villages, evaluate and explore the individual qualities and heritage values of a cluster of traditional dwellings in western Hunan, and rate the exterior quality of individual traditional dwellings, thus systematically revealing how indicators affect the value of traditional homes [52].Landscape conservation experts use digital technology [53], such as satellite remote sensing [54], drone photogrammetry [55], multi-baseline digital close-up photogrammetry [56], constructed a digital detection and early warning system for traditional village landscapes, and efficiently processed, computed and analysed through networked big data, involving traditional village heritage conservation [57], landscape mapping [58], virtual tourism [59,60].In the field of village tourism, Luh Ketut Yulitrisna Dewi, in exploring the tourism development model of Pancasari, suggests that local wisdom is a factor that local governments, investors and communities must take into account when developing a sustainable tourism model for villages [61].Moslem Ghasemi, who studied the appropriate tourism model for two villages in the Lower Kinabatangan region from a stakeholder perspective, pointed out that the development of a new tourism model should take into account the local people's perceptions and expectations of the tourism business in their villages while listening to the voices of the villagers [62].Joo Y et al. used the theory of planned behaviour (TPB) model to determine the impact of the use of social networking services (SNS) on village tourism to help sustainable tourism promoters and marketers use the results to make more effective decisions [63].
In China, the exploration of traditional village landscapes spans multiple disciplines, including human geography, landscape ecology, history, and anthropology.This interdisciplinary approach has yielded a wealth of research findings in several key areas, such as village landscape planning [64,65], the ecological functions of landscapes [66], the cultural values inherent within them [67,68], and the development of tourism resources [69].For instance, Liu Peilin first put forward the concept of cultural landscape genes', marking a pioneering step in the systematic study of the landscape genes of traditional villages [70,71].Another example is to construct the concept of "corridor, patch, and matrix" to evaluate the classification of areas through the ecological effect of the rural habitat environment [72].From the perspective of "aesthetic historical geography", it is pointed out that primitive nature worship is the main psychological motivation for the generation of rural cultural landscapes [73].Some scholars have also carried out landscape perception analyses for specific villages.Zhuang Qianda et al. used NVivo 12 to analyse and qualitatively code the content of comments on social media platform websites about Yagou village and studied the landscape characteristics of Yayou Gou village [74].Weng Feifan et al. used Python (3.12.0) to collect the digital footprint evaluation of tourists, used ROST CM6 to sort out the general village landscape features of the visit perceivers, and finally constructed a multidimensional landscape perception model of Xiti Village.
The overview reveals that contemporary research on traditional villages concentrates on physical spatial forms, ecological symbiosis, and tourism patterns.However, there is a noticeable gap in the systematic protection of the overall landscape environment of tourist-oriented traditional villages.This oversight neglects the crucial role that documenting landscape features plays in the future planning of traditional village landscapes.

Research Methodology
This study draws on and incorporates the Zhuang QD [74] and Weng FF [75] approaches to landscape perception research.The research employs a kind of scientific approach, beginning with the collection and cleaning of web evaluation texts from nine tourism websites using the Octopus data Collector.This dataset undergoes analysis to uncover tourists' perceptions of Xijiang Thousand Households Miao Village's landscape, using tools such as word frequency and sentiment analysis.Insights from this analysis inform a deeper exploration of the village's landscape features using Nvivo12 for coding.The study concludes by identifying strengths and weaknesses in the village's landscape through semantic network and grounded theory coding, leading to targeted improvement strategies.As shown in Figure 1.

Word Frequency Analysis
The frequency of use of words in a certain range of linguistic material is known as word frequency [76].Word frequency analysis is a traditional content analysis method in bibliometrics.The basic principle is to determine hot spots and their changing trends by counting the frequency of vocabulary appearances and counting and analysing the number of occurrences of important vocabulary, which is an important means of text mining.
Since the word frequency analysis method is based on objective data, it has relatively high accuracy, and it can reduce the influence of personal subjectivity on the analysis results and enhance the credibility of the analysis, so it has been widely used in various fields [77].Therefore, the meticulous refinement and analysis of high-frequency words play a crucial role in systematically unravelling travellers's perceptions of the landscape of Xijiang Thousand Households Miao Village.

Semantic Networks and Outgoing and Incoming Degree Analyses
Semantic Web is an application of complex web technology in the field of natural language processing, that expresses and stores natural language through custom nodes and relation types [78].The structure of knowledge is usually represented by an interconnection pattern of nodes and lines [79].Each node represents a keyword, and the co-occurrence relationship between keywords and keywords is described as the internal constituent relationship and its structure in a certain domain, which enables the visual analysis of the attributes between the evaluation objects [80].The denser the network of paths surrounding a node, the more integral its lexical content is to the web of related elements, highlighting its significance in shaping the tourist experience.In such a directed graph, the in-degree is represented by the number of arrows pointing to each keyword, signifying its influence or importance, while the out-degree is denoted by the arrows emanating from the keyword to other nodes, illustrating its connectivity.Semantic network analysis, especially in the realm of communication, exhibits remarkable flexibility.It adeptly navigates through disparate snippets of information from varied sources, unveiling their underlying linkages.Thus, by employing a semantic network graph alongside a detailed analysis of in and out degrees, we can effectively discern the structural relationships among the high-frequency words pertaining to the landscape features of Xijiang Thousand Households Miao Village, revealing a comprehensive picture of its perceived landscape.

Sentiment Analysis
Sentiment analysis, a specialised branch of text analysis, is categorised into coarsegrained and fine-grained sentiment analysis based on the granularity of the classification subject.Its primary goal is to extract sentiment information related to specific aspects mentioned within the text.In the context of this study, sentiment analysis serves as a crucial tool to gauge tourists' satisfaction with the open space landscape system.This assessment acts as a vital benchmark to determine whether the landscape system of Xijiang Thousand Households Miao Village succeeds in delivering a positive experience to its visitors.

Rooted Theory Coding Analysis
Rootedness theory, first proposed by Strauss scholars [81,82], is a qualitative research method that combines systematicity and flexibility to construct a theoretical system by rooting empirical data and following rigorous research procedures.Researchers construct theoretical generalisations from raw data by coding, conceptualising, and categorising qualitative data.This process yields a theoretical framework intricately linked to the studied phenomenon, allowing for the progressive abstraction of the research object's core characteristics.Such a framework not only elucidates but also predicts.In this particular investigation, the QSR Nvivo 12 analytical tool facilitated the coding process across three stages, enabling the identification of landscape feature types within Xijiang Thousand Households Miao Village.
1. Open coding serves as the foundational step for bottom-up theoretical modeling.At this stage, researchers approach the data with an open mind, aiming to identify as many potential concepts and categories as possible.They then proceed to name and preliminarily interpret these findings, ensuring the data are thoroughly deconstructed and conceptualised.
2. Spindle coding is that after identifying a series of concepts, the researcher needs to identify the main concepts among them and build up their attribute dimensions around these core concepts, searching for associations between different concepts and forming a network of concepts to analyse the correlations that exist between different relationships.3.After correlating the various concepts, the researcher needs to choose a core concept, and then establish the systematic connection and theoretical framework between this core concept and other concepts in a top-down manner, forming a typical relational structure based on the main category [83].The main category is based on a typical relational structure.

Study Area
In this research, the Xijiang Thousand Households Miaozhai, located in Nangui Village, Xijiang Town, Leishan County, Qiandongnan Miao, and Dong Autonomous Prefecture, Guizhou Province, China, is selected as the study area.The Xijiang Miao Village is an ethnic tourist attraction in the eastern line of Guizhou, which covers an area of about 5.5 square kilometres (Figure 2), with 1468 households and more than 6000 people living in the village, of which the Miao ethnic group accounts for 99.5% of the total population [84].The reasons for selecting this area for this research are as follows: 1. Xijiang Thousand Households Miao Village is well preserved and has a long history.
Xijiang Thousand Households Miao Village is one of the most characteristic and wellpreserved traditional ethnic villages in China, with a history spanning more than 2000 years.It was listed as a provincial cultural relics protection unit in Guizhou Province in 1992.The village's architecture, terraced fields, and traditional customs reflect a more comprehensive and typical rural humanistic and natural landscape.2. Xijiang Thousand Houses Miao Village is unique in its ethnic characteristics.Xijiang Thousand Households Miao Village, as the largest Miao settlement village in China at present [85], is the production and living space of the Miao people and carries the functions of tourism and public service at the same time, which is valued by the local government and is the key demonstration area of the traditional village.3. Xijiang Thousand Houses Miao Village attracts a large number of domestic and foreign tourists.The average daily passenger flow of bookings in the off-season (November-April) is 10,000, while in the peak season (April-November), it can be as high as 100,000, with a hundred-fold increase in tourists in just ten years, generating more than 10 billion yuan of tourism revenue, making it an extremely popular representative attraction of the characteristic traditional villages [86].Therefore, tourists leave a large number of comments on major online platforms, providing a rich data base for this study.

Data Collection and Processing
In this study, supported by online comments, we collected 24,096 visitor review text data about Xijiang Thousand Households Miao Village as of 1 November 2023, using the Octopus Data Collector with the above nine tourism websites as the data source.In order to ensure that the data are reasonable and the scope of the landscape-centred study is not deviated, after collecting the review data, we made the necessary pre-processing, and the specific criteria for screening included: 1. Reviews of travel products with strong recommendations have been removed, and their content usually includes the name of the shop, its location, and the merchant's contact information.2. Invalid comments that contain only emoticons, numbers, and garbled code, and comments of less than three characters that cannot be determined to be descriptive of landscape features.3. Duplicate comments from the same user caused by the technical network are selected and deleted.4. The correctness and legality of the data are checked, and both foreign and traditional Chinese texts are converted into simplified Chinese texts.
Based on the above data processing process, a total of 587,884 invalid characters were excluded, and 13,691 valid data were obtained and saved to the valid database.Table 1 demonstrates the comparison between the original data collected from each platform and the amount of comment data after secondary screening.

Text Evaluation Data Characteristics
The frequency of words is closely related to the experience of tourists, and the high-frequency words reflect that the factors referred to by this term receive higher attention and play a more important role in the formation of the sense of place.Using the ROST CM6 (5.8.0.603) software, the top 50 keywords were intercepted from the online comment data and sorted from highest to lowest, resulting in a high-frequency word table (Table 2) and a high-frequency word cloud (Figure 3), which contains 13 verbs, 74 nouns, and 13 adjectives.The words with the highest frequency involve tourist destinations, administrative divisions, and local representative landscapes.The high-frequency words can be divided into the following categories: landscape features words (nouns), including specific physical areas and landscapes that tourists focus on perceiving, such as "Night landscape", "performance", "nature", "museum", and so on.The emotional perception words, such as "worthy" and "passionate", mainly focus on expressing the emotional feelings of tourists after experiencing Xijiang Thousand Households Miao Village.The environmental perception words ("adjectives"), mainly express the tourists' perception of the overall environment of Xijiang Thousand Households Miao Village, such as "natural" and "commercial".Play behaviour ("verbs") can reflect the main behaviours and purposes of tourists in Xijiang Thousand Households Miao Village.The word "sightseeing" appears as frequently as 1183 times, and "taking pictures" appears 1097 times, which shows that taking pictures is the core purpose of most tourists in Xijiang Thousand Households Miao Village.

Evaluating the Interconnectedness of Words
According to the method described in section 3.1, we obtain the semantic network relationship map of high-frequency words (Figure 4) and the values of out-degrees and indegrees (Table 3) of the evaluation elements of tourists' perceptions of Xijiang Thousand Households Miao Village, in order to reflect the linkage and hierarchical relationship of high-frequency words.Among them, the core circle words "Miao village", "scenic spot", and "commercialisation" are located in the core position with the highest relevance in the network, representing the basic source of tourists' landscape perception.Secondly, words such as "viewing platform", "Night view", and "scenic area" are in the sub-core position, reflecting visitors' further understanding of the landscape system of Xijiang Thousand Households Miao Village.The rest of the content extending outward from the core circle is loosely connected, mainly tourists' specific views on various landscape elements.

Profiling of Public Emotional Attitudes
Table 4 shows the results of the sentiment analysis of 13,691 tourists, which can be classified into three levels of positive, neutral, and negative sentiment gradients.From the results of the analysis (Table 5), the positive emotions are 10,248, dominated by 74.86%, followed by the negative emotions, accounting for 24.02%, and the neutral emotions are the least, accounting for only 13.70%.It can be seen that tourists' evaluation of Xijiang Thousand Households Miao Village is mostly positive emotions, predominantly presented in the natural landscape and ethnic and humanistic landscape, while the negative emotions are concentrated in the service and management, which is of strong reference significance for Xijiang Thousand Households Miao Village in the subsequent protection and utilisation as well as in the planning and management.

Landscape Features Coding Results
The results of the landscape feature coding identification are detailed in Table 6.This analysis builds on previous classifications, notably the landscape fan classification [74,75], through the extraction and amalgamation of 208 distinct codes.These were systematically categorised into 28 subcategories to delineate the secondary landscape features of Xijiang Thousand Households Miao Village.Subsequently, these subcategories were consolidated into five primary categories that collectively encapsulate the village's predominant landscape features.These categories are: natural landscape (n = 441, 2%), cultural and social landscape (n = 10,630, 48.3%), seasonal landscape (n = 74, 0.3%), perceptual landscape (n = 7368, 38.5%), and services and infrastructure (n = 3499, 15.9%).Furthermore, a strong causal relationship was identified between user reviews and landscape experience, as documented in Table 7. Through the analysis and synthesis of the underlying logic of the original commentaries, a structural model was developed to encapsulate the factors influencing user evaluations at Xijiang Thousand Households Miao Village.As depicted in Figure 5, the visual, social, and cultural aspects predominantly shape the landscape experience, while the natural landscape, seasonal landscape, and services and infrastructure serve as secondary influencing factors.Natural Landscape → Landscape Experience The regional environment reflects the ethnic characteristics, constitutes the secondary landscape features of the Miao village, and indirectly affects the landscape experience.
Xijiang Thousand Households Mziao Village is located in a typical river valley, the clear Baishui River through the village.On the slope of the northeast side of the river, thousands of years of hard-working and brave Xijiang Miao compatriots have lived here for sunrise farming, sunset farming, and rest.The main body of Miao village is located on the slope of the northeast side of the river.

Perceptual landscape → Landscape Experience
Phenomenon perception expresses the national spirit, constitutes the main landscape features of Miao stockade, and directly affects the landscape experience.landscape experience.
The ancient village minority customs do not want to abandon the convenience brought by modern society, with a flat mood to the Xijiang small rest, wind and rain bridge to sit, see grandma singing embroidery,and at night in the river clear bar to drink a small wine, rain bridge to sit, see grandma singing embroidery, at night in the river, clear bar to drink a small wine, happy.

Seasonal Landscape → Landscape Experience
Seasonal changes restrict the shape of the scenic spots, constitute the secondary landscape features of the Miao village, and indirectly affect Seasonal changes restrict the shape of the scenic spot, constitute the secondary landscape features of the Miao village, and indirectly affect Winter is still too cold, there are too many people, and we need an off-season.We lived with the locals for a few days in order to feel the real culture.
Services and Infrastructure → Landscape Experience Tourism facilities reflect the consciousness of scenic spot, constitute the secondary landscape features of Miao village, and indirectly affect the landscape experience.
The scenic spot is still quite distinctive, but the whole tour experience is very poor, and the service level of the staff in the scenic spot is extremely The whole tour experience is very poor, and the service level of the staff at the scenic spot is extremely poor.
Obviously, it can be found in Table 6 that the cultural and social landscape contains a total of 12 secondary landscapes, which are all mentioned by tourists 10,630 times in all 9 tourism softwares, which fully proves the core position of Landscape features elements related to cultural and social elements in tourists' perceptions.They specifically include: historical landscapes (Figure 6a), cultural elements (Figure 6b), architectural landscapes (Figure 6c), festival and event-type landscapes (Figure 6d), food and beverage landscapes (Figure 6e), and night landscapes (Figure 6f), as confirmed by the comments of several tourists: "Miao ethnic group's original ecological song and dance performance, the beautiful West River Miao song and dance evening, viewing platform, to see the song and dance performance you must go to the scene, the live performance is really too beautiful.""Twelve barriers to wine, the story must have started with wine, with mountains, water, style, wine, and beautiful stories."Secondly, there were 7369 comments related to perceptual landscape, accounting for 33.5%, second only to cultural and social landscape.After coding, the Xijiang Thousand Households Miao Village Perceptual Landscape Free Codes can be divided into three categories: environmental features, spiritual connotations, and perceptual features.As far as environmental features are concerned, tourists' evaluation texts recorded the subject's perception process and preference tendency of the corresponding landscape perception object from multiple dimensions, such as the selection of attraction objects, the language of describing the picture, and the actual feeling of the scene.For example:

"A very pretty place,I would like to visit next time, with mountains, beautiful scenery and history."
"The weather here is not too hot, it's cooler, the spiritual mountains are beautiful, and the people are simple.""The air is so fresh in the morning, and the village is a peaceful scene with wisps of green smoke hanging from the eaves of every house."Tables 8 and 9 illustrate the polarisation of tourists' attitudes in the category of spiritual qualities: 444 tourists agreed that Xijiang Thousand Households Miao Village is a place of ethnic minorities; 211 tourists thought that Xijiang Thousand Households Miao Village has maintained the original flavour; and 113 tourists thought that Xijiang Thousand Households Miao Village is a place of cohesion in the Miao culture.However, some tourists still regretted the development of Xijiang Thousand Households Miao Village.2328 tourists mentioned "commercialisation" and "commercial atmosphere" in 9 different apps; 127 tourists mentioned "homogenisation" in seven apps, including Meituan.Negative comments such as "lack of interaction" also appeared from time to time.The appearance of the Perceptual Features category evaluation also confirms the landscape value of Xijiang Thousand Households Miao Village.It is a very commercial scenic spot, which is far away and difficult to row by ferry, and the walking distance is very long.The feeling of taking photos everywhere is difficult to rise 5A.overtourist 9 844 15.30 percent There are too many people, especially when trying to take a photo.Almost everywhere is packed especially on the water bridge, it is impossible to squeeze in.homogenisation 7 127 2.30 percent Not bad, but the homogenisation is too serious.There is no particularly attractive place.
Large-scale photography base 5 99 1.70 percent Scenic spots do not say generalisation, all light shows which are nothing special.Felt like a photography base, where everyone is taking photos.Good locations are occupied Scenic spots do not say generalisation.Shopping centre 2 2 0.30 percent Two steps from a shop, three steps from a hotel, directly across the Guzhai shopping street.
Finally, our findings indicate that the current landscape planning of Xijiang Thousand Households Miao Village aligns well with tourists' actual landscape perceptions.Table 10 compiles all announcements from the "5A Creation" section on the official website of Xijiang Thousand Households Miao Village as of 1 November 2023, which were extracted and categorised to assess their core content.This analysis reveals that the village's development policy has consistently focused on social culture, economy, and tourism services, reflecting a strong emphasis on the cultural and social landscape.These findings corroborate our earlier results.However, although this top-down planning mode coordinates and plans the development of Xijiang Thousand Households Miao Village as a whole, it may neglect or even eliminate some other landscape features, such as natural landscape perception.

Relationship between Landscape Characters and Visitor Perceptions
As emphasised in the European Landscape Convention, "to comprehensively and systematically describe landscape features, it is inevitable that the whole process of the user's participation and experience is indispensable.post-perceptual evaluation".The data from online comments can promptly correct the bias of managers in updating village landscape areas and also provide new perspectives for other traditional village landscape management work.This result is consistent with Carneiro et al [96].Carneiro argues that earlier studies of landscape in rural tourism relied too heavily on visual stimuli from information such as photographs, which are not entirely reliable.From the results compiled in this study, it is clear that tourists do not stay in a particular landscape for a long time while visiting the village, nor do they follow the route set by the scenic area manager completely.From Xijiang Thousand Households Miao Village's key cultural ritual landscapes, such as "xijiang ceremony", "gala party", and "wine tasting ceremony", to the service facility landscapes, such as "stone slab path", "mountain path", and "wine tasting ceremony", tourists do not stay in one landscape for a long time, nor do they follow the route set by the scenic area manager.The fact that the landscape features were mentioned many times by the tourists, such as the "stone road", "mountain path", and "tourist bus", shows that the tourists' grasp and perception of landscape features are much more detailed than the visual stimuli obtained by the researcher through photographs.In fact, the direct contact between tourists and landscapes can produce a large number of unquantifiable experiences, and tourists can obtain pleasurable, nostalgic, and stimulating feelings by focusing on heterogeneous humanistic and natural landscapes in the process of travelling.If tourists are immersed in the landscape, they can leave pleasant memories, and at the same time, they will be affected by the facilities and basic services when experiencing the landscape, so the development of the landscape will be more sustainable by referring to the feedback of the tourists' impressions after the experience.

Intrinsic Relationships between the Five Types of Landscape Characters
In terms of intrinsic relationship, cultural and social landscape, visual landscape, and natural landscape are interconnected and interact with each other (Figure 7), and together they constitute the main body of the landscape of Xijiang Thousand Households Miao Village, but these three are often limited by scenic services and infrastructure at the same time.From the overall relationship: all landscape elements are affected by seasonal changes and the perceptual landscape of tourists.With the change of seasons, these landscape elements will show a different style.In addition, with the seasonal changes, the landscape space and visitors' perceptions will also change.Not only does this support the visitor's perspective of Tudor C's view [97] that elements of nature, culture, perception, and aesthetics are at the core of landscape features distinction, but it also confirms Callau's et al. [98] study of the landscapes of the Ebro Delta Natural Park, a protected area in Spain, mentioning that facilities and basic services are one of the important attributes that distinguish it from the analysis of the content of the photographs.

Core Elements in the Formation of Visitors' Perception
Through data coding, we found that cultural and social landscape was mentioned a total of 10,630 times in nine social software, accounting for 48.30% of the total ratio, which is the core factor influencing the Xijiang Thousand Households Miao Village visitors' perception.However, previous studies on Xijiang Thousand Households Miao Village have all focused only on the key position of tourists in the study of Xijiang Thousand Households Miao Village and have concentrated more on the characteristics of the settlement habitat [99], ethnic and cultural resource conservation [100], ethnic tourism [101], and traditional houses [102].In the cultural and social landscape of Xijiang Thousand Households Miao Village, "night landscape", "architectural landscape", and "traditional houses", the important role of various landscape features in tourists' perceptions has been neglected.Among the cultural and social landscapes of Xijiang Thousand Households Miao Village, the three landscapes of "night landscape", "architectural landscape", and "dietary landscape" were particularly concerned by tourists, and were mentioned 2569 times, 2183 times, and 1987 times, respectively.2569, 2183, and 1987 times, respectively.
Combined with Table 9, it can be found that the three, as the focus of the Xijiang Thousand Households Miao Village tourist attractions, are characterised by open-ended experience, multi-sensory stimulation, and situational substitution, which also confirms Chen Xi's [103] conclusion constructed through PLS-SEM.In addition, the word frequency analysis shows that "night landscape", "food", "footstools", and so on are the high-frequency words that have been repeatedly mentioned in the tourists' comments, and this symbolic landscape element is the most important element of the tourist attraction.This symbolic landscape element [104] is also one of the core elements in the formation of tourists' perceptual intention towards tourism purpose, but the public seems to have a weak perception of the cultural connotations of the landscape, which is in line with the findings of Jingjing Zhou [105].

Issues for Improvement in Village Tourism Development
In the process of compilation, we also found that tourists' comments not only identify the village landscape features, but some of the comments reflecting negative emotions also expose the problems that need to be improved in the development of village tourism.From the description of the network text, we get: 1.The natural landscape elements and seasonal landscape elements of Xijiang Thousand Households Miao Village are far less developed than the cultural and social landscape elements.Therefore, at the level of landscape planning and design, there is an urgent need for managers to formulate strategies and guidelines for the use of natural landscape under the protection of the existing good conditions, such as combining the natural landscape with the activity landscape, exploring the landscape composite, and activating the experience economy on the basis of respecting nature through the production of local food, agricultural experience, and other folklore experiential activities.2. The road system and the charging system of Xijiang Thousand Households Miao Village is much criticised.The stone road is not friendly to tourists with luggage, and the chaotic car park layout and charging system bring many troubles to tourists.Therefore, more consideration should be given to the appropriateness and practicality of the facility system for the human body in public facilities, and humanistic care should be reflected in the management level of the scenic spot.3. The over-commercialised development of Xijiang Thousand Households Miao Village has made the cultural genes gradually disappear.Therefore, at the level of spiritual and cultural construction, we should apply benign commercial development to promote the dissemination of the rural cultural kernel.For example, reasonable pricing of ethnic cultural activities, reducing blindly catering to the market in ethnic song and dance performances, and showing more original ethnic rituals and activities.

Over-Commercialisation of Scenic Spots Brings a Series of Problems and Coping Strategies
As a representative of the traditional village in tourism, Xijiang Thousand Households Miao Village has been included in the national "scenic area with village" tourism poverty alleviation demonstration project.The rapid development of tourism will inevitably have a certain degree of negative impact on village society and the natural environment, such as the gentrification of the village [106,107].The development and construction of traditional village landscapes through rural tourism is a reconstruction of village physical space against the background of gentrification.For traditional villages undergoing tourism, the sustainability of the landscape and the unity of management are considered to be the keys to promoting the sustainable development of rural tourism and culture [108].In terms of the results of the sentiment analysis in this paper, the negative sentiment of Xijiang Thousand Households Miao Village focuses on the excessive commercialisation of the scenic area.What tourists expect to experience is the original Miao culture and rural style, while the urbanised commercial street' form is widely given in the village space.The production of new public space in Xijiang Thousand Households Miao Village is often oriented towards commercial layout, and the living space of the indigenous people also makes concessions for commercial development.The official development of public space is guided by tourism, and the publicity often emphasises the commercial and entertainment aspects.At present, the landscape of A has already seen the phenomenon of cultural "symbolisation" and "performativity", with "ethnic clothing travel photography" and homogenised "Netflix" shops almost everywhere.The collective rituals based on folk beliefs have completely evolved into profit-making "tourist shows", losing the solemnity of folk rituals (Figure 8).Excessive commercialisation has led to the loss of a sense of place in the countryside [109].As a result, tourists' travel experiences are affected, and it is difficult to develop tourism in traditional villages in a high-quality and sustainable way.In view of the above problems, from the perspective of village landscape planning, tourist-type traditional villages should pay more attention to the construction of human life order, actively construct local life scenes, and incorporate the indigenous people's life scene into the construction of the landscape so as to restore the originality of the traditional village landscapes, such as the construction of the public space for the villagers to communicate and interact with each other, which not only enhances the villagers' connection with society but also maintains the emotional connection of the local communities and follows the original logic of life and history of the villages.It not only follows the original logic of village life and historical lineage but also makes foreign culture more compatible with local culture.On the other hand, to tap into the excellent traditional culture of the countryside, we should rely on the original ethnic culture for cultural and tourism integration and form local characteristics such as ethnic performing arts, ethnic cuisine, ethnic costumes, traditional skills, special inns, special cultural creations, etc.We should also reduce the proliferation of "performative" folk activities, organise folk rituals on specific ethnic festivals to reflect the living style of the ethnic culture, and enhance the compatibility of foreign culture with local culture.The living style of ethnic culture can enhance the "original ecological" image of ethnic villages.At the same time, the development process should balance the interests of various parties, carry out reasonable planning and layout of rural tourism resources, guide the distribution of differentiated business forms, and avoid homogenised vicious competition.In the face of excessive "commercialisation", we can try to formulate a sustainable tourism development convention from the perspective of the statute by the government, scenic spot development companies, villagers and village collectives, etc., to control foreign capital, housing rental and sale, new facilities, etc., and tilt the distribution of benefits to the original inhabitants who are not renting or selling houses and engaging in agricultural production to form a standardised and effective landscape governance model to promote the development of traditional rural tourism resources and to promote the development of traditional rural tourism.effective landscape governance model to promote the sustainable development of traditional village tourism, which fits the findings of Weili Yang [110].

Conclusions
This study utilises the framework of a tourist-centric traditional village to explore the landscape perceptions of visitors.Specifically, Xijiang Thousand Households Miao Village in Guizhou Province serves as the case study site.Employing a combination of online review data and analytical software tools such as ROST CM6 and NVivo 12, this research extracts high-frequency terms related to landscape perceptions from tourist reviews.These terms are then used to construct a semantic network, facilitating the analysis of tourists' affective responses to both the quality of the landscape and the level of services provided at the site.The study successfully identifies five primary types of landscape features and delineates a structural model comprising 28 sub-categories of landscape features.
It is found that direct contact between tourists and scenic landscapes generates a large number of experiences that are difficult to quantify, and the impression feedback from these experiences is important for enhancing the sustainability of landscape development.At the same time, cultural and social landscapes, visual landscapes, and natural landscapes in scenic areas are constrained by scenic services and infrastructure, and seasonal variations can also have a significant impact on visitor experiences.Among the five types of landscape features, cultural and social landscapes play an important role in enhancing tourist satisfaction as the core factors influencing tourists' psychological perceptions.It is worth noting that symbolic landscape elements are one of the core elements in the formation of perceptual intention; however, tourists' perception of the cultural connotations embedded in the landscape is still weak, which suggests that we should strengthen the excavation and display of cultural elements in landscape construction.In terms of sentiment analysis, most tourists hold positive sentiments towards Xijiang Thousand Households Miao Village, but there are also some negative sentiments, mainly focusing on the relatively weak natural landscape, the unsatisfactory road system and charging system, and the over-commercialised development.In response to these problems, this study puts forward corresponding planning suggestions.In summary, this study improves the methodology of landscape characterisation using tourists' online evaluation texts and enriches the research on landscape characterisation of traditional villages of the same type of tourism.Meanwhile, the results of this study reflect the accelerated trend of the deep integration of culture and tourism in traditional villages in China, respond to the policy guidance of protecting traditional villages, traditional houses, and historical and cultural villages and towns' under the policy of China's rural revitalisation, and provide useful references and insights for the landscape construction and development of the Xijiang Thousand Households Miaozhai and even similar tourist-type traditional villages.
However, this study is not without its limitations.First, compared to traditional qualitative data collection methods such as interviews and field surveys, the online review data used here may suffer from variability in writing quality and may not fully capture the authentic perceptions and thoughts of tourists.Further exploration into more effective data filtration methods is necessary to achieve more objective results.Second, this study overlooked the visual data from pictures included in the comments, which contain valuable spatial features not easily discernible in text.Future research should expand the scope of data collection to include these visual elements, perhaps by collecting and uniformly analysing comments alongside their accompanying images and converting these into textual descriptions for comprehensive analysis.These limitations will be explored and resolved in further research.

Figure 1 .
Figure 1.Research methodology and research steps.

Figure 3 .
Figure 3. Word cloud of online review data from travel websites.

Figure 4 .
Figure 4. Semantic network analysis of online review data on travel websites.

Figure 5 .
Figure 5. Structural model of landscape features of Xijiang Thousand Miao village.

Figure 6 .
Figure 6.Landscape features related to cultural elements, social elements in Xijiang Thousand Households Miao Village: (a) historical landscapes; (b) cultural elements; (c) architectural landscapes; (d) festival and event-type landscapes; (e) food and beverage landscapes; (f) night landscapes.

Figure 7 .
Figure 7. Interrelationships of Landscape Features of Traditional Villages.

Figure 8 .
Figure 8.The commercialised landscape of Xijiang Thousand Households Miao Village: (a) landscape facilities in square; (b) commercial street form; (c) ethnic clothing travelling photography; (d) tourism exhibition real life map.

Table 1 .
Comparison of quantities before and after data processing.

Table 3 .
High-frequency words semantic network out and in-degree values.

Table 4 .
Three gradients of results based on sentiment analysis.

Table 5 .
Positive and negative emotion segmentation statistics.

Table 6 .
Open coding of landscape features.

Table 7 .
Selective coding results: implicit logic of core categories.

Table 8 .
Categories of positive mental qualities.Especially like the garden scenic spot, the water is clear to the bottom, children and adults can play in the water, the water is not deep, the bottom has pebbles.As soon as you step into Thousand Households Miao Village in Xijiang, you feel like you are in an open-air museum that showcases the rich history and traditional culture of the Miao people.

Table 9 .
Categories of negative mental qualities.

Table 10 .
Tweets and announcements in the "5A Creation" section.