Changes in Social Media Big Data on Healing Forests: A Time-Series Analysis on the Use Behavior of Healing Forests before and after the COVID-19 Pandemic in South Korea

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This article examined changes in visitor behavior and visitor interest in healing forests before and after the COVID-19 pandemic in South Korea, mainly based on data extracted from two domestic portal sites NAVER and DAUM. It is of practical significance but needs some improvements. Please see my comments and suggestions below:

1. The usage behaviors reflected by the two portal sites may be biased, simply because not all visitors to healing forests would use these portals. The authors should discuss this and admit it as a limitation. 

2. It leads to biases when T1 covers two years, while T and T2 cover only one year as shown in Table 2. The span of T, T1, and T2 should be equal, i.e., all cover one year or two years, to allow comparison. 

3. Topics in Table 3 need some manual adjustments based on CONCOR analysis. For instance, keywords related to NRF should be excluded as they are irrelevant, and the most suitable category for each keyword should be carefully examined and should better be consistent across periods. It is confusing as "barrier-free" is under the Visitation and usage behavior of T while under Healing Forest and NRF of T1, and "healing" is under Camping of T1 while under Healing forest and NRF in other periods. Usage differences and shifts between different periods revealed in Table 3 should be discussed in-depth to draw some major conclusions.    

4. Discussions should be rooted in results and previous literature, while conclusions should be extracted from discussions. Currently, "Therefore, visits to healing forests have been affected by the COVID-19 pandemic, shifting into individual- and small group-centered visits" in lines 262-264, and "As healing forests are located in forests, the demand and behavior of visitors change depending on the season. Forest managers should provide programs that take into account seasonality and expand activities outside the forest by linking with cultural and tourist resources near the forest" in lines 303-306 are not supported by data and results in this study. The authors should reconsider these carefully and involve references in the section of the Discussion. 

5. The authors claimed that "These findings will contribute to the development of future marketing strategies and programs for healing forests". This is not convincing right now as the current/ historical marketing strategies and programs for healing forests are not reviewed. The reviews of healing forests in lines 95-105 need to be reorganized and supplemented to directly support the topic and focus of this article. 

6. Typos should be carefully checked and corrected, e.g., "xx" in line 31, missing information in Table 2, "Table 6" in line 220, etc. 

Author Response

Review #1

Thank you for your thoughtful comments.

 

1. The usage behaviors reflected by the two portal sites may be biased, simply because not all visitors to healing forests would use these portals. The authors should discuss this and admit it as a limitation.

             ☞ The comments were taken into account and written as limitations of this study and ways to improve future research. (Line 342~350)

 

2. It leads to biases when T1 covers two years, while T and T2 cover only one year as shown in Table 2. The span of T, T1, and T2 should be equal, i.e., all cover one year or two years, to allow comparison.

             ☞ For time-series analyses using text mining, there are studies with different time periods (Reference 25, 40). Therefore, we have added the studies with different time periods to the text. (Line 140~143)

 

3. Topics in Table 3 need some manual adjustments based on CONCOR analysis. For instance, keywords related to NRF should be excluded as they are irrelevant, and the most suitable category for each keyword should be carefully examined and should better be consistent across periods. It is confusing as "barrier-free" is under the Visitation and usage behavior of T while under Healing Forest and NRF of T1, and "healing" is under Camping of T1 while under Healing forest and NRF in other periods. Usage differences and shifts between different periods revealed in Table 3 should be discussed in-depth to draw some major conclusions.

             ☞ The results of CONCOR analysis are grouped between closely related nodes (words), which means that highly related words are grouped together according to the usage behavior of users by time period, and even if the same word appears in each time period, the group changes according to the relationship of the user's usage behavior. "barrier-free" is affected by this change in usage behavior and is now in a different group for each time period. We've added a section to the body about this. (line 227~230)

 

4. Discussions should be rooted in results and previous literature, while conclusions should be extracted from discussions. Currently, "Therefore, visits to healing forests have been affected by the COVID-19 pandemic, shifting into individual- and small group-centered visits" in lines 262-264, and "As healing forests are located in forests, the demand and behavior of visitors change depending on the season. Forest managers should provide programs that take into account seasonality and expand activities outside the forest by linking with cultural and tourist resources near the forest" in lines 303-306 are not supported by data and results in this study. The authors should reconsider these carefully and involve references in the section of the Discussion.

             ☞ We've made some general changes to the discussion to reflect your feedback.

 

5. The authors claimed that "These findings will contribute to the development of future marketing strategies and programs for healing forests". This is not convincing right now as the current/ historical marketing strategies and programs for healing forests are not reviewed. The reviews of healing forests in lines 95-105 need to be reorganized and supplemented to directly support the topic and focus of this article.

             ☞ The focus of this study is to understand the behavior and interests of visitors to the Healing Forest before and after COVID-19, and to identify factors that influenced the changes. The part you mentioned can be seen as the implications of the paper, as the factors that changed (tourism, hiking) were identified.

In addition, domestic healing forest research is limited to identifying the effects of healing programs; therefore, it is difficult to reconstruct and supplement the previous research in lines 95~105 with marketing/program aspects. To complement this, a future research need was mentioned that marketing/programs should be based on the identified healing forest change factors. (Line 307-309)

 

6. Typos should be carefully checked and corrected, e.g., "xx" in line 31, missing information in Table 2, "Table 6" in line 220, etc.

             ☞ Fixed to reflect your comments. (Line 31, 225)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper is good. Please improve it based on the below-described comments:

1. Line 31: "of xx on". What does "xx" mean here?

2. Line 72: Table 2 should be referred before Table 1.

3. "under Article 2". What doest this mean?

4. The programs/moudels used for your analysis should be described in more detail, and references with the program versions should be presented. The methodology used in the paper is not very common, so deeper descriptions would be beneficial for readers. I mean especially the Textom program here.

5. Table 2: Please use "T" as an abbreviation for pre-COVID-19 period. You used "T" in other places in tha paper, so it seems that this "T" should be presented here too.

6. Table 4: p-value less than 0.001 should be written as "p < 0.001", because p-value can not equal 0.000.

7. "Across T, T1, and T2, we found no significant difference in the top 254 trending words". How the statistical differences were revealed? What tests did you used for testing these? See also other places in the paper with the similar cases.

8. I found only one reference in the discussion section. I believe that the discussion should be expanded, and other relevant references should be provided.

Author Response

Review #2

Thank you for your thoughtful comments.

 

1. Line 31: "of xx on". What does "xx" mean here?

             ☞ I deleted the "xx" part from the body due to an error in translation.

 

2. Line 72: Table 2 should be referred before Table 1.

             ☞ Table 2 is placed below Table 1 because it supplements the part of the analysis period described from Line 132.

 

3. "under Article 2". What doest this mean?

             ☞ Means the "Act on the Promotion of Forest Welfare" in table 1.

 

4. The programs/moudels used for your analysis should be described in more detail, and references with the program versions should be presented. The methodology used in the paper is not very common, so deeper descriptions would be beneficial for readers. I mean especially the Textom program here.

             ☞ We've added a description of the Textom program based on your comments. (Lines 74-79)

 

5. Table 2: Please use "T" as an abbreviation for pre-COVID-19 period. You used "T" in other places in tha paper, so it seems that this "T" should be presented here too.

             ☞ Fixed to "T" to reflect your comment. (Table 2)

 

6. Table 4: p-value less than 0.001 should be written as "p < 0.001", because p-value can not equal 0.000.

             ☞ Corrected to p<0.001 based on feedback. (line 255)

 

 

7. "Across T, T1, and T2, we found no significant difference in the top 254 trending words". How the statistical differences were revealed? What tests did you used for testing these? See also other places in the paper with the similar cases.

             ☞ Due to the nature of the data (text) used in text mining techniques, it is difficult to statistically determine the similarity of the words themselves. Therefore, to determine the statistical similarity, we created a matrix based on 71 common words per period in the existing submission files T, T1, and T2, and performed QAP correlation analysis to determine the statistical similarity of the network. We believe that the question wants to statistically show the differences in words across time periods. Therefore, we re-created the matrix with 100 words per time period, ran the QAP correlation analysis, and re-drew the results to supplement the discussion and conclusion section. (Lines 251~268)

 

8. I found only one reference in the discussion section. I believe that the discussion should be expanded, and other relevant references should be provided.

             ☞ Added references based on your comments.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This study aims to identify changes in visitor behavior and visitor interest in healing forests before and after COVID-19 in Korea. This theme is relevant and interesting from both the perspective of human health and the ecological standpoint because it can influence governmental institutions' interest in these environments.

 

This work is very clear but too concise in some parts.

 

a)     The introduction is well done and clear, but it could be enhanced by expanding upon the sources cited and including more international comparisons.

 

b)     I have just few observations to make regarding the methodology because it is well explained and supported by accompanying materials.

 

2.1 : Line 84:

Please explain in which law is found this : Article 2 of the Act on the Promotion of Forest Welfare. And add the reference.  

 

Additionally, it would be beneficial to describe more characteristics of Korean healing forests in comparison to other forests. For example, are they managed regularly? What is their size? Where they are? Perhaps add a map.  How do healing forests differ from other types of forests (e.g. different plant composition etc.)? Providing such details would enrich the understanding of the unique features and attributes of Korean healing forests and it would be useful in the discussion.

 

c)     The discussion section requires improvement in two key areas.

Firstly, it lacks comparison with other studies, which could provide valuable insights. Therefore, it should be expanded to include a discussion of the results obtained in relation to findings from other works.

Secondly, the potential change in the use of healing forests, mentioned by the authors as a possible increase in tourism and attraction, is not adequately addressed. Several pertinent questions could be explored in this context. For instance, how might this change affect the function of healing forests? Is there a risk of increased tourist activity causing damage to the forest ecosystem, or are there measures in place to mitigate such risks?

Furthermore, could the increased use lead to the expansion of healing forests? These questions, among others, should be addressed to provide a comprehensive analysis of the potential implications of changing usage patterns on healing forests.

 

Additionally, the limitations of the research are not adequately discussed but are merely mentioned in the Conclusions section.

 

d)     The References list requires revision. In some cases, the date is repeated, and there are inconsistencies in the use of italics. These errors should be corrected for clarity and consistency.

Author Response

Review #3

Thank you for your thoughtful comments.

 

1. The introduction is well done and clear, but it could be enhanced by expanding upon the sources cited and including more international comparisons.

             ☞ Expanded the journal to reflect feedback, and added and corrected partial references related to international comparisons. (Line 50)

 

2. 2.1 : Line 84 > Please explain in which law is found this : Article 2 of the Act on the Promotion of Forest Welfare. And add the reference.

Additionally, it would be beneficial to describe more characteristics of Korean healing forests in comparison to other forests. For example, are they managed regularly? What is their size? Where they are? Perhaps add a map. How do healing forests differ from other types of forests (e.g. different plant composition etc.)? Providing such details would enrich the understanding of the unique features and attributes of Korean healing forests and it would be useful in the discussion.

             ☞ The tree species planted inside the healing forests are cypress, which secretes a lot of phytoncides, as well as pine and pine forests. Currently, there is a lack of research on the spatial characteristics of healing forests in Korea, which is an important clue to reveal the healing mechanism of forest healing (Jung and Shin, 2023). Therefore, we have written a section related to the law, which is a clear expression in the existing file.

Additionally, we have created maps related to the location and status of healing forests in the country. (Line 94, Appendix A)

 

• Jung, K.M. and W.S. Shin. The Relationship between Perceived Importance of Space and Users’ Satisfaction. Korean Journal of Environment and Ecology 2023. 37, 273-288.

1. The discussion section requires improvement in two key areas.

Firstly, it lacks comparison with other studies, which could provide valuable insights. Therefore, it should be expanded to include a discussion of the results obtained in relation to findings from other works.

Secondly, the potential change in the use of healing forests, mentioned by the authors as a possible increase in tourism and attraction, is not adequately addressed. Several pertinent questions could be explored in this context. For instance, how might this change affect the function of healing forests? Is there a risk of increased tourist activity causing damage to the forest ecosystem, or are there measures in place to mitigate such risks?

Furthermore, could the increased use lead to the expansion of healing forests? These questions, among others, should be addressed to provide a comprehensive analysis of the potential implications of changing usage patterns on healing forests.

Additionally, the limitations of the research are not adequately discussed but are merely mentioned in the Conclusions section.

             ☞ We have revised the discussion in general in light of the reviewers' comments and other research findings. In addition, we believe that Korea's healing forests are currently 38 evenly distributed across the country, and that there is a need for qualitative rather than quantitative growth.

In addition, the Forest of Healing is operated on a reservation system to consider the ecological carrying capacity of the forests within the Forest of Healing and to provide quality program services. Therefore, we believe that the potential for damage to the ecosystem within the Forest of Healing due to increased tourist activities is unlikely to be significant.

 

4. The References list requires revision. In some cases, the date is repeated, and there are inconsistencies in the use of italics. These errors should be corrected for clarity and consistency.

             ☞ Revised the reference list to reflect your feedback.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

This revised manuscript is much improved than the last version. However, there are still a few issues that need to be addressed.

1. The authors discussed a lot on difference in usage patterns between T and T2. It is better to base on difference tests, instead of correlation analysis, to get firm results. If the difference between T and T2 is not statistically significant, the discussion and conclusions need to be reconsidered. 

2. The layout of the results in Table 4 need to be corrected. The figures in Table 4 also need careful double-checking as they are significantly different from those in the previous version. 

3. In text citations should start from 1 and be numbered consecutively as they appear. 

Author Response

Review #1

Thank you for your thoughtful comments.

 

1. The authors discussed a lot on difference in usage patterns between T and T2. It is better to base on difference tests, instead of correlation analysis, to get firm results. If the difference between T and T2 is not statistically significant, the discussion and conclusions need to be reconsidered.

            ☞ In the case of text mining, the mainstream of research has been to create a matrix of words themselves to discover their "relationships " (Refs. 24, 25, 39, 40). This is because, due to the unstructured nature of text, the relative correlation between unique words in an entire sentence can be quantified, but a single unique word cannot be quantified by assigning a number to it. Therefore, while it is possible to statistically identify items that are related, it is not possible to identify 'differences'. Therefore, in this study, we propose a different approach to correlation analysis.

 

2. The layout of the results in Table 4 need to be corrected. The figures in Table 4 also need careful double-checking as they are significantly different from those in the previous version.

            ☞ The results layout in Table 4 is based on previous research (Reference 24) and was carefully reviewed by the authors.

 

3. In text citations should start from 1 and be numbered consecutively as they appear. 

            ☞ We've made changes throughout to reflect your feedback.

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for your considerable edits. Please find some issues for reconsidering:

1. Lines 342-350: These are limitations. Please create a separate section for limitations. In general, conclusions are too long, therefore, eliminating limitations from conclusions would be beneficial.

2. Practical implications should be also presented in a separate section. I feel that restructuring this will make your paper more impactful and easy to read.

3. My previous comment ("p-value less than 0.001 should be written as "p < 0.001", because p-value can not equal 0.000") was ignored. I would like to ask the authors twice to reconsider the paper acc. to this comment.

4. Please use two decimal places (not three) for presenting correlation coefficients, as it is common practice in science.

5. Line 88: Please add a reference for this Article 2.

Author Response

Review #2

Thank you for your thoughtful comments.

 

1. Lines 342-350: These are limitations. Please create a separate section for limitations. In general, conclusions are too long, therefore, eliminating limitations from conclusions would be beneficial.

               ☞ Based on your feedback, we have organized section 4.2. Implications and limitations and removed the limitations from the conclusions. (line 328)

 

2. Practical implications should be also presented in a separate section. I feel that restructuring this will make your paper more impactful and easy to read.

               ☞ Based on your feedback, we've organized and revised the 4.2. Implications and limitations section to include limitations. (line 328)

 

3. My previous comment ("p-value less than 0.001 should be written as "p < 0.001", because p-value can not equal 0.000") was ignored. I would like to ask the authors twice to reconsider the paper acc. to this comment.

               ☞ We' ve changed the p < 0.001 to "p < 0.001" when submitting the revision to reflect your feedback, please double-check. (line 257)

 

4. Please use two decimal places (not three) for presenting correlation coefficients, as it is common practice in science.

               ☞ In text mining analysis, three decimal places are used to represent the results of correlation analysis (Wang et al., 2024; Yu et al., 2023); therefore, we also used three decimal places in this study.

- Wang, W.; Mao, W.; Wu, R.; Zhu, J.; Yang, Z. Study on the Spatial Imbalance and Polarization of Marine Green Aquaculture Efficiency in China. Water 2024, 16, 273. https://doi.org/10.3390/w16020273

- Yu, A.; She, H.; Cao, J. Evolution of the Spatial Patterns of Global Egg Trading Networks in the 21 Century. Sustainability 2023, 15, 11895. https://doi.org/10.3390/su151511895

 

5. Line 88: Please add a reference for this Article 2.

               ☞ Originally written as the Law on the Promotion of Forest Welfare, the detailed description of the facilities is given in another law (Law on Forest Culture and Recreation), so we have revised it. (line 88~92)

We have also added the law to the reference 18 in response to feedback.