Cultural Perception of Tourism Heritage Landscapes via Multi-Label Deep Learning: A Study of Jingdezhen, the Porcelain Capital

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article provides a comprehensive overview of recent research methods, demonstrating excellent organization and addressing all aspects of a research study and ensures that all crucial components of a research study are covered in a comprehensive way.  It is an extensive overview of the research methodologies The conclusions and the efficacy of this study in understanding or implementing strategies for mass tourism are not evident. Since the Special Issue focuses on mass tourism, this study should approach the topic comprehensively. It should include evidence demonstrating how the study's findings can address relevant tools for local heritage management and planning.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I apologize for the fact that I judge the article to be of exceptionally good quality. The researchers use a new approach, so the content of the article may be of wide professional interest. The research methodology is interesting, logically structured and of high quality. Another important point is that the method can be used by others in a similar environment. Both the quantity and the depth of the included literature is suitable for the foundation of the research and the results.

Congratulations to the authors.

Author Response

Thank you for your encouraging words and acceptance of our manuscript. We greatly appreciate your positive feedback.

Reviewer 3 Report

Comments and Suggestions for Authors

Comments:

1. Row 110: "Deep learning technologies" – The authors likely meant to say "Deep learning techniques" instead.
2. Figure 2. Overview of the Jingdezhen region – The maps have a coordinate grid and are oriented north, so the north arrow is redundant and should be removed.
3. Row 212: "ResNet-50 is a deep neural network consisting of 50 layers" – This statement requires at least one citation.
4. Rows 244-245: "Therefore, we incorporated the following improvement strategies into the ResNet framework." – The authors likely refer to adaptations to the study's objectives, as Multi-information Feature Fusion (MFF), Spatial Attention Module (SAM), and Channel Attention Module (CAM) are fundamental components in deep learning, even though they were not originally part of ResNet.
5. Rows 462-463: "Furthermore, these cultural perception levels were categorized into five grades: low, medium-low, medium, medium-high, and high." – How was this classification established? The same classification levels appear in reference [13] cited by the authors, suggesting a potential methodological overlap that should be clarified.
6. Figure 10. Cultural recognition results of 11 CCHL in Jingdezhen – The legend uses ‘cultural perception,’ while the caption refers to ‘cultural recognition’. It is necessary to clearly define the concept of ‘cultural perception’ within the study's framework.
   • Are these two concepts considered equivalent?
   • This is a sensitive issue, and the audience must understand the meaning of cultural perception in the context of deep learning to avoid misinterpretations.
   • Additionally, the filter introduced by selecting three ceramic artists to construct the annotation dataset (rows 277-279) should be addressed. The psychosensory filtering applied by these three individuals inherently introduces subjectivity into the evaluation.
7. Rows 478-480: "In order to examine the spatial variations in ceramic cultural type perception among different cultural landscapes, this research utilized ArcGIS natural breaks methodology to perform quantitative analysis on cultural type distribution within 11 CCHL areas."
   • Natural Breaks (Jenks) is not a methodology exclusive to Esri software – it is a widely used classification scheme in GIS and RS.
   • What specific Esri software version was used? ArcGIS Pro or an older desktop version?
   • The classification method identifies threshold values within a dataset, but it is unclear what numerical data was used for the natural breaks classification.
8. Rows 488-489: "In the displayed chart, columns a and b represent the absolute quantities and 488 relative quantity characteristics of cultural perception respectively."
   • Figure 11 is not a chart but a set of maps.
   • The north arrow is unnecessary, and the content is illegible, so the maps need to be redone.
9. Subsection 4.2. Quantification of Cultural Diversity in CCHL and Figure 12.
   • Were the same Natural Breaks classification thresholds applied to the two indices (Shannon Diversity Index and Simpson Index Distribution)?
   • Typically, this classification method is used in the initial phase of research, when values are not yet well-defined, and then the class boundaries are reinterpreted. Was this step followed in the analysis?

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

Article entitled “Cultural Perception of Tourism Heritage Landscapes via Multi-Label Deep Learning: A Study of Jingdezhen, the Porcelain Capital” is focused on an interesting research topic. The manuscript is centred on the analysis of the cultural perception of typical heritage landscapes in Jingdezhen, a renowned historic cultural city, a historical porcelain center, utilizing a mixed methodology (social media big data analysis combined with ResNet-50 deep learning model and ArcGIS spatial analysis).

 

Key words: the authors used several keywords that are also in the title: landscape tourism heritage; cultural perception; Multi-label classification; Deep learning; Jingdezhen. Typically, keywords should not be similar; the authors could use instead Jingdezhen - historical center of porcelain; they can use as keywords production culture; Cultural diversity; spatial analysis,

Introduction

The authors should present sinthetically more general details about the topic: e.g. why the topic is important; the authors can related (sinthetically) the porcelain production with the values of industrial heritage; the introduction of an article has the scope to present sintheticaly the topic in general.

The research questions should be presented after the presentation of aim of the manuscript. The authors should underline the importance of the study.

Literature review

2.1. Cultural Heritage Landscape Tourism

The authors should add 2 or 3 citations (Rows 66-69):  Since 1970, Western countries have continuously innovated cultural heritage  tourism models, promoting its development into a popular form of tourism. This form o tourism attracts visitors by transforming historical and cultural assets into commodities [7, here],  playing crucial roles in promoting cultural exchange, local economic development, and heritage preservation [8,9 and here].

Please insert at least one example for each issue enumerated (Rows 71-73): Numerous scholars have conducted in-depth research on cultural heritage tourism from various perspectives, accumulating rich theoretical findings in tourist experience [here],  cultural heritage preservation [here], and tourism development [here]. You ca nadd also authors from European countries. You used mostly studies that are focsed on the ealuation of the culture heritage in China.

2.3. Deep Learning and Multi-Label Classification Methods

At the end of this sub-section the authors include details about the methodlological aspects (Rows 136-143). This fragment should be presented at the methodology section. It is a subsection of literature review! Moreover, the first figure should be included in the methodology section.

Methodology

3.1. Study Area

In this fragment you should add the citation for statistical data (Rows 158-160): From January to August 2024, the city received 41.6245 million domes- 158 tic tourists, an increase of 18.97% year-on-year, with total tourism revenue reaching 47.868  159 billion yuan, up 13.76% year-on-year.

3.5. ArcGIS Spatial Analysis

To include at least a citation for the presentation of the method (Rows 398-401):  In spatial analysis, cluster analysis is a commonly used technique that focuses on grouping data points based on specific similarity characteristics. The purpose of this technique is to partition data points into natural groups, where points within each group exhibit high similarity, while points between different groups are relatively distinct.

Results and discussion

I missed a discussion of results. The results of the study should be analyzed critically and compared with results obtained in other studies focused on similar methodology or with similar case studies located in different countries to highlight similarities or differences.

4.4. Theoretical Contributions and Practical Significance to Tourism Heritage Landscape Planning 

(1) Theoretical Contributions: the authors should reflect the novelty of the study.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

In figure 10, correct in the map `Sculpture procelain Factory`.

Author Response

We thank the reviewer for their careful examination of our figures. The spelling error in Figure 10 has been corrected，

changing 'Sculpture Procelain Factory' to 'Sculpture Porcelain Factory' in the map.

Please see the attachment.

Author Response File: Author Response.pdf