Resident Effect Perception in Urban Spaces to Inform Urban Design Strategies

Round 1

Reviewer 1 Report

The authors of this manuscript shed light on quantifying emotional expression of the crowd in public spaces with developed computational graphical models and analyzed their correlations with physical environmental elements. Finally, the authors use the percentage of these elements with high correlation to positive emotions as a reference to evaluate design schemes of public spaces.

 

I appreciate the efforts of the authors in integrating multiple tools into a proposed design workflow for improving the spatial qualities of urban spaces. Some of them such as the facial expression recognition may indeed have the potential of taking over the traditional questionnaire surveying in studying human perceptions to public spaces.

 

However, the above-described highlights of this manuscript were hard to receive because this manuscript was poorly formulated for following reasons: 1) the title was too generic around the topic, not guiding audience into the topic; 2) the introduction spoke too generically about the shifts of focuses on urban design and machine learning models. But it did not reach down precisely to the point, where the study was built on and aimed at. 3) the definitions of terms were often absent or unprecise throughout the manuscript. Certain terminologies illustrated in the tables or graphics were even not mentioned in the main text. 4) The statements about the research questions and objectives were unclear. 5) Certain key information was totally missing in the manuscript, for example, the labels for the semantic segmentation models, meaning of EE_i values, the further use of TE_i values. 6) The conclusion is made full of unprecise descriptions, not really summarizing up the results and contribution of this study.

 

Another drawback of this manuscript was the misuse of chi-square tests in section 5.1. This was methodologically wrong. I recommended to use standard deviation against the reference value for the author’s purpose. It must be stated in detail how the expectation values as the references were calculated.

 

 

Please find my comments on individual lines below:

  

@multiple places through the manuscript:

Delete all “meticulously”; the considerations in selecting software and making certain decisions should be described. It was not helpful to the audience when this word replaced the actual reasoning process.

citizen -> resident; Citizen tends to specify a person who legally belongs to a country, and resident is used, generally, for a person who is legally living or working in a particular locality.

Citation formatting is wrong at multiple places through the introduction session. MDPI journals use ACS Style.

 

@Line 98: the introduction in section 1.2 is off the points. In the manuscript, the relevant machine learning tools are image semantic segmentation and image face recognition. Please specify the relevant models on the points instead of speaking generically.

 

@Line 142: This section somehow contains the most important arguments for your research questions and hypothesis. However, they are not well formulated. I would suggest highlighting the gaps in your proposed design workflow first (see comments at line 198), and state research questions to the audience accordingly which gaps you are bridging.

 

@Line 149: Which articles? Please add citations.

 

@Line 198: This figure is very hard to interpret. I would recommend 1) matching each step with the sections in the manuscript. 2) Highlight the gaps (problems) you concluded in section 6.

 

@Line 200: The figure title is not methodological framework, but a proposed design workflow.

 

@Line 202: the way you formulated the objectives are not easy to follow. An objective could be, for example, developing a novel method, verifying the application of a certain procedure, or making a new discovery about something. Your statements were like a to-do-list.

 

@Line 211: lacks an overview of the experimental purpose and aim, what data you would like to collect, and an overall process. Without these guides, readers have a hard time following various methods described later.

 

@Line 231: lacks an explanation of why the chosen site suits the study purpose.

 

@Line 234: unclear how the units are set on the satellite maps, is it a manual decision?

 

@Line 234: Second -> Secondly

 

@Line 237: What does colors of the grids represent? Please add a legend.

 

@Line 263: How to ensure that the image contains at least 10 participants? What about the non-participants passing by?

 

@Line 278: please add citation to this software and the model. It is unclear if you used a trained model offered by this software? Or did you use this software for training your own model? In case it is an exist model, please specify how these labels (sky, ships, structures…) fits your aim.

 

@Line 285: What objects does “Structures” in figure 4 refer to? What are the numbers representing?

 

@Line 290: Add citation to the software ENVI. Same as the image recognition software, please specify which model you are using. Is it a built-in model or did you train your own model.

 

@Line 298: What are the grids at the bottom? I can only guess now they are one grid unit as an example to show the classification result. In this case, I would suggest zooming in this grid also at the satellite image and the middle classification image. Please also match the grid order with the legend.

 

@Line 308: add citation to this software.

 

@Line 324: it is not explained how the crowd size was counted.

 

@Line 326: what is crowd mixture?

 

@Line 334: it is not well explained what data you used for the age and gender recognition. Did you record videos/photos at multiple locations? How to remove duplicates? It must also be clarified regarding the deviation of this recognition method.

 

@Line 361: “4”->“3”. Please indicate if the equation is built on previous studies or is your own “creation”. In the second case, it should be clarified why you use equal weight on different environmental descriptors for calculating the total impact. It was not clearly illustrated where this TE_i was used in further analysis.

 

@Line 370: Effects -> Emotional Effects? all people -> the crown / the participants

 

@Line 379: What are “urban spacery” and “boats”? They are not included in the planar elements in 3.5.2.

 

@Line 381: this conclusion is not supported by figure 9. I suggest referring to figures 14 and 15, where you report the planar elements of different grids.

 

@Line 383: not explained well what the EE value was calculated from the six scores in emotional analysis.

 

@Line 385: Effects -> Emotional Effects? Is it still about the emotional effect in section 4.2 or total effects?

 

@Line 397: Figure 10 has no names for x axis and has no y axis. The group names are not the same as the age groups described in the text. For discrete space units, I cannot see a reason for this curved graphic style. The white curve at the bottom is not understandable. It does not refer to any age group.

 

@Line 389: group photo and graphic design elements are undefined.

 

@Line 394: what is “urban farscape”?

 

@Line 395: “dotted tree” undefined.

 

@Line 399: Some elements remain undefined in the text such as “boys, girls, animals”. The numbers are not readable in the main text. You should either use a bigger font or split each sub figure into individual ones. I would also suggest considering moving these sub figures to supplementary material. Why the elements for correlation analysis are different for different age groups?

 

@Line 403: the title of table 1-5 are all the same. What are these tables? What does “distribution position” mean? How is the shape of the element acquired? The descriptions are all undefined in the text like “Faceted”, “Ribbon” etc.

 

@Line 459: The phrases “causing” and “triggered” were used in this paragraph, which is not precise. This study, in my understanding, analyzed the correlation between variables in multiple aspects and emotional facial expressions. There is no further evidence to support casualties between them.

 

@Line 460: “hard element” undefined.

 

@Line 465: the percentage of different planar elements are doubled in the top bar graphic and the middle table. In the table, these elements were called “plane element” while they were named planar elements in earlier text. Artificial and natural elements were not indicated in the bar graphics at the bottom.

 

@Line 472: It is not illustrated how the space units were selected for summarizing the male’s and female’s facial expressions. How did the different scenes lead to a fair comparison. Why not use the same space units?

 

@Line 501: diffe-rences -> difference

 

@Line 501: When using the Chi-square test, the variables must be mutually exclusive. The spatial elements are apparently not exclusive in the scenarios. In this section, therefore, Chi-square test makes completely none-sense. In a very generous estimation, the author may aim at quantifying the effects of the design schemes. For this purpose, this study may use standard deviation of proportions of spatial elements against a target value acquired in this study. Even though, how these target values were calculated was poorly illustrated in this manuscript.

 

@Line 513: “5”->”4”

 

@Line 519: “6”->”5”

 

@Line 549: Accuracy of the element recognition, facial recognition must be addressed in this section.

 

@Line 580: delete “this study holds great significance and value for 580 both society and the design field, as”

 

@Line 584: “external and internal pressures” undefined

I was not criticizing about the grammar, but the terminologies. Many of them used in the manuscript are not precise. This caused barriers in delivering the contents. I would recommend this manuscript going through a language check service after the revision.

Author Response

Response to Reviewer 1 Comments

 

Thank you for your professional insight into this paper and for your comments on revisions.

Based on your suggestions, we summarized the following 46 modifications:

 

1. Delete all “meticulously”; the considerations in selecting software and making certain decisions should be described. It was not helpful to the audience when this word replaced the actual reasoning process.

Response 1:

In response to this suggestion, lines 238 of the thesis were modified as follows:

” …are classified into five distinct categories…”

Lines 260 of the thesis were modified as follows:

“…have been curated…”

Lines 302 of the thesis were modified as follows:

“…was employed to facilitate…”

Lines 554 of the thesis were modified as follows:

“…a venue crafted to attract…”

 

2. citizen -> resident; Citizen tends to specify a person who legally belongs to a country, and resident is used, generally, for a person who is legally living or working in a particular locality.

Response 2:

 

In response to this suggestion, lines 2 of the thesis were modified as follows:

” Resident Effect Perception in Urban Spaces to Inform Urban Design Strategies”

Lines 24 to 25 of the thesis were modified as follows:

” …residents. By taking into account the specific driving relationships between each element and residents…”

Lines 31 to 32 of the thesis were modified as follows:

” …with residents.”

Lines 33 of the thesis were modified as follows:

“Resident effect”

Lines 37 of the thesis were modified as follows:

“Transformation of Urban Design Towards Resident Emotion Orientation”

Lines 62 of the thesis were modified as follows:

“…of residents by…”

Lines 65 of the thesis were modified as follows:

“…improve residents' sense…”

Lines 146 of the thesis were modified as follows:

“…account residents' perceptions…”

Lines 146 of the thesis were modified as follows:

“…incorporating residents' inherent…”

Lines 151 of the thesis were modified as follows:

“…and residents presents…”

Lines 158 of the thesis were modified as follows:

“…analyze residents' emotions…”

Lines 169 to 170 of the thesis were modified as follows:

“…effects on residents in external surroundings. The subjective assessment of residents' emotions relates…”

Lines 176 of the thesis were modified as follows:

“…characteristics of residents who…”

Lines 178 of the thesis were modified as follows:

“…elements and resident…”

Lines 179 of the thesis were modified as follows:

“…residents.”

Lines 449 to 450 of the thesis were modified as follows:

“…emotions on residents of …”

Lines 535 of the thesis were modified as follows:

“…effects on residents …”

Lines 537 of the thesis were modified as follows:

“…effects on residents and…”

Lines 542 of the thesis were modified as follows:

“…hearts of residents.”

Lines 618 to 619 of the thesis were modified as follows:

“…we collected 470 residents samples as quantitative data for resident structure and emotions.”

Lines 623 to 624 of the thesis were modified as follows:

“…between residents and their cities.”

Lines 628 of the thesis were modified as follows:

“…capture resident emotions and explore the relationship between residents and urban…”

Lines 633 to 640 of the thesis were modified as follows:

“…of residents, urban spaces, and emotions by employing on-site research, analysis, and mapping of the element composition and proportion rules of the Yangpu Riverside public space demonstration section. The study utilized space elements, type classification, and the capture of residents' emotions to establish a foundation for future space design research. Secondly, it empirically explored the relationship between urban space elements and residents' preferences and emotions. The results indicate that a well-defined urban space design strategy can assist governments in attracting specific groups and establishing better communication with residents.”

Lines 652 of the thesis were modified as follows:

“…and residents' structures…”

 

3. Citation formatting is wrong at multiple places through the introduction session. MDPI journals use ACS Style.

Response 3: In response to this suggestion, the thesis were modified to use ACS Style.

 

4. Line 98: the introduction in section 1.2 is off the points. In the manuscript, the relevant machine learning tools are image semantic segmentation and image face recognition. Please specify the relevant models on the points instead of speaking generically.

Response 4:

In response to this suggestion, lines 103 to 124 of the thesis were modified as follows:

“The underlying principle of deep learning is to combine low-level features to create more abstract highlevel representations, such as attributes, categories, or features, to discover the distributed feature representation of data[17,18]. Deep learning has been extensively applied in the field of computer vision as a potential way to have ac-cess to graphic information and to perform visual challenges [19]. Semantic segmenta-tion is a fundamental computer vision tool that requires pixel-by-pixel prediction of a given image. Remarkable progress in semantic segmentation has been made in recent years with the development of deep neural networks (DNNs) [20-23]. For instances, Tencent AI lab has improved the Cityscapes [24] semantic segmentation mIoU by 56.3% and 53.0% by utilizing DeepLab-v2 network and ResNet-101 backbone[25]. In the meantime, face recognition, as one of the most common computer vision tasks, has been extensively studied for decades [26-28]. The primary procedure of face recognition is face verification and face feature discrimination [26]. In recently years, in an attempt to address the disadvantages of CNNs and traditional face recognition methods in terms of weak feature discrimination, some new methods have been introduced, such as ArcFace, SphereFace, CosFace, and it has been demonstrated that the CosFace method has achieved state-of-the-art performance in face recognition[29]. Benefiting from the accelerated advancement and integration of intelligent technologies, human body information and 3D spatial composition can be identified more accurately. Accordingly, there are researches on deep learning numerical processing functions based on image segmentation and face recognition to obtain hidden knowledge about the design from experimental and computational data to guide the design [30].”

 

5. Line 142: This section somehow contains the most important arguments for your research questions and hypothesis. However, they are not well formulated. I would suggest highlighting the gaps in your proposed design workflow first (see comments at line 198), and state research questions to the audience accordingly which gaps you are bridging.

Response 5:

In response to this suggestion, lines 190 to 207 of the thesis were modified as follows:

“In general, to tackle the foregoing research gaps, we proposed a new approach that combines traditional data collection methods and emerging technologies to quantify crowd use preferences for pedestrian public spaces in urban clusters. Further, based on the quantitative results, Space, People, and Emotion (SPE) model was constructed with the aim of applying the interaction between residents' emotions and spaces obtained in this study to the evaluation of urban design proposals, and providing novel design methods and concepts for urban space design. This study comprised three stages. The initial stage involved preparatory research, where the study area was examined qualitatively through literature and online resources. Subsequently, the space research plane scale was established and divided into modules for sequencing. The second stage entailed the quantitative study of the space, with data collected during the target period. Two databases were employed as data sources, namely Google satellite map images and on-site research data. This stage involved cleaning, categorizing, and analyzing the collected data and the effect map, emotional effect map, age-subdivided effect map, and space rules are obtained. Finally, the SPE model was applied to the evaluation of really urban design solutions  (Figure 1).”

 

6. Line 149: Which articles? Please add citations.

Response 6:

In response to this suggestion, lines 165 of the thesis were modified as follows:

“Despite numerous articles [42-47]that offer…”

lines 165 of the thesis were modified as follows:

“42. Marin, J.; De Meulder, B.J.J.o.L.A. Urban landscape design exercises in urban metabolism: reconnecting with Central Limburg's regenerative resource landscape. 2018, 13, 36-49.

43. Ingaramo, R.; Voghera, A. Topics and Methods for Urban and Landscape Design; Springer: 2016.
44. Zou, M.; Zhang, H.J.E.S.; Research, P. Cooling strategies for thermal comfort in cities: a review of key methods in landscape design. 2021, 28, 62640-62650.
45. Liu, M.; Nijhuis, S.J.E.I.A.R. Mapping landscape spaces: Methods for understanding spatial-visual characteristics in landscape design. 2020, 82, 106376.
46. Mahmoud, A.H.; Omar, R.H.J.F.o.A.R. Planting design for urban parks: Space syntax as a landscape design assessment tool. 2015, 4, 35-45.
47. He, M.; Wang, Y.; Wang, W.J.; Xie, Z.J.I.J.o.G.; Parks. Therapeutic plant landscape design of urban forest parks based on the Five Senses Theory: A case study of Stanley Park in Canada. 2022, 10, 97-112.”

 

7. Line 198: This figure is very hard to interpret. I would recommend 1) matching each step with the sections in the manuscript. 2) Highlight the gaps (problems) you concluded in section 6.

Response 7:

In response to this suggestion, lines 208 of the thesis were modified as follows:

“ “

 

8. Line 200: The figure title is not methodological framework, but a proposed design workflow.

Response 8:

In response to this suggestion, lines 209 of the thesis were modified as follows:

“Figure 1. Flowchart of this research”

 

9. Line 202: the way you formulated the objectives are not easy to follow. An objective could be, for example, developing a novel method, verifying the application of a certain procedure, or making a new discovery about something. Your statements were like a to-do-list.

Response 9:

In response to this suggestion, lines 211 to 215 of the thesis were modified as follows:

“1) A multi-dimensional research system of 10*10 small-scale spatial units has been constructed;

2) Established the research method of emotional influence law of space on age segmentation and gender segmentation by combining experience and artificial intelligence;

3) A goal-oriented program evaluation system for emotions is implemented.”

 

10. Line 211: lacks an overview of the experimental purpose and aim, what data you would like to collect, and an overall process. Without these guides, readers have a hard time following various methods described later.

Response 10:

In response to this suggestion, lines 218 to 225 of the thesis were modified as follows:

“This study applied a combination of field research and AI analysis to perform the methodology. Prior to the field study, we first downloaded a Google satellite plan of the site, divided the site into 10*10m units and labeled them with serial numbers. After entrancing into the site, we acquired images of the space and participants based on field photography. Based on the images taken in the field, we applied image analysis methods to obtain spatial element information and participant characteristics, and supervised learning methods to obtain site plan element information. The data obtained from the above analysis served as the basis for the regularity study.”

 

11. Line 231: lacks an explanation of why the chosen site suits the study purpose.

Response 11:

In response to this suggestion, lines 103 to 124 of the thesis were modified as follows:

“Yangpu Riverside Public Space is a strategic development area in the 14th Five-Year Plan for the development of "One River, One River" in Shanghai, China, and it is an "initial place" for the significant concept of the People's City. The site of this study is the demonstration section of Yangpu Riverside Public Space, which plays an important role as a model for the whole Riverside Public Space and provides an effective reference significance.”

 

12. Line 234: unclear how the units are set on the satellite maps, is it a manual decision?

Response 12: Thank you for your comment regarding the setting of units on the satellite maps. In our study, we manually divided the site based on the current scale base of the site. As shown in Figure 2, the width of most of the local plots is about 10m, leading us to create a 10m*10m cell as the study unit. The process of defining these units is crucial for our data analysis, as it allows us to segment the urban space into manageable units for further study. In our case, a 10m*10m cell appears to be a reasonable choice considering the scale of the local plots.

 

 

13. Line 234: Second -> Secondly

Response 13:

In response to this suggestion, lines 253 of the thesis were modified as follows:

“Secondly, any duplicates were…

 

14. Line 237: What does colors of the grids represent? Please add a legend.

Response 14:

In response to this suggestion, lines 256 of the thesis were modified as follows:

“”

 

15. Line 263: How to ensure that the image contains at least 10 participants? What about the non-participants passing by?

Response 15: Thank you for your insightful comments and queries. In our study, we define "participants" as individuals who visit the park with the intention to engage with the space, such as for recreation, rest, or social interaction. On the other hand, "non-participants" or "passersby" refer to those who traverse the park without actively engaging with it. Since this area functions as a park rather than a transportation space, we considered that most people come to this area with the purpose of visiting or resting, so this paper defaults to the people who come to this space as space participants.To ensure that the image contains at least 10 participants, we focused on data collection during peak visiting hours. We agree with your point that passersby do exist, and their interactions with the space can differ from those of participants. This is an important aspect that we will further distinguish and explore in future studies to refine our understanding of park usage patterns.

 

16. Line 278: please add citation to this software and the model. It is unclear if you used a trained model offered by this software? Or did you use this software for training your own model? In case it is an exist model, please specify how these labels (sky, ships, structures…) fits your aim.

Response 16:

In response to this suggestion, lines 298 to 305 of the thesis were modified as follows:

“Semantic segmentation is effectively able to provide pixel-level category information. Ln this paper, we applied Semantic Segmentation to recognize spatial elements [59]. The methodology used in Tencent's Semantic Segmentation technology achieved state-of-the-art performance on challenging benchmarks, significantly out-performing previous methods with Mean Intersection over Union (MioU) up to 55.1% [25]. Tencent Image Recognition Free Online Trial Platform 2015 was employed to facilitate batch recognition of a vast collection of space images, resulting in the identification of distinctive element tags and their corresponding image proportions (Figure 4).”

 

17. Line 285: What objects does “Structures” in figure 4 refer to? What are the numbers representing?

Response 17:

In response to this suggestion, lines 310 of the thesis were modified as follows:

“”

Lines 574 of the thesis were modified as follows:

“”

Lines 580 of the thesis were modified as follows:

“”

 

18. Line 290: Add citation to the software ENVI. Same as the image recognition software, please specify which model you are using. Is it a built-in model or did you train your own model.

Response 18:

In response to this suggestion, lines 314 to 322 of the thesis were modified as follows:

“Supervised classification methods were considered to be relatively more precise in achieving quantitative categorization of remotely sensed data [60,61]. In this study,we applied supervised classification methods to classified the Google Earth satellite images of the site. ENVI software 5.3 Region of Interest (ROI) tools was employed for the comprehensive recognition and cluster analysis for supervised classification. We utilized the Support Vector Machine (SVM) algorithm within the ROI tool because it performed relatively better in extracting urban areas[62].

We created nine new ROI and selected multiple geometric sample ranges (geometry) for each land type ROI as training samples for supervised learning.”

 

19. Line 298: What are the grids at the bottom? I can only guess now they are one grid unit as an example to show the classification result. In this case, I would suggest zooming in this grid also at the satellite image and the middle classification image. Please also match the grid order with the legend.

Response 19:

In response to this suggestion, lines 103 to 124 of the thesis were modified as follows:

“”

 

20. Line 308: add citation to this software.

Response 20:

In response to this suggestion, lines 344 of the thesis were modified as follows:

“…software Extreme Vision Free Web Trial Platform 2015…”

 

21. Line 324: it is not explained how the crowd size was counted.

Response 21:

In response to this suggestion, lines 361 to 363 of the thesis were modified as follows:

“…Crowd size data was the number of results exported by Tencent face recognition software (the method is shown in the following section 3.5.5).…”

 

22. Line 326: what is crowd mixture?

Response 22: Thank you for your question regarding the term "crowd mixture." In the context of our study, "crowd mixture" refered to the diversity of populations present in a given urban space. This diversity could encompass a variety of factors including age, gender, among others. Understanding the crowd mixture is crucial in our study as it provides insights into the diverse ways different age and gender groups interact with the urban space, their preferences, and their needs. This information is valuable for urban planners and designers in creating inclusive spaces that cater to the diverse needs of all users.

 

23. Line 334: it is not well explained what data you used for the age and gender recognition. Did you record videos/photos at multiple locations? How to remove duplicates? It must also be clarified regarding the deviation of this recognition method.

Response 23:

In response to this suggestion, lines 374 to 378 of the thesis were modified as follows:

“Tencent face recognition model was an effective face recognition tool including the age and gender recognition, achieving 99.80% accuracy at LFW 2017 [29]. We first applied the face recognition model on the Tencent Free Online Trial Platform 2015 to recognize the character data in the filmed video, then manually cleaned the repetitive data, and finally obtained the age and gender data of the participants.”

 

24. Line 361: “4”->“3”. Please indicate if the equation is built on previous studies or is your own “creation”. In the second case, it should be clarified why you use equal weight on different environmental descriptors for calculating the total impact. It was not clearly illustrated where this TEi was used in further analysis.

Response 24:

In response to this suggestion, lines 406 to 407 of the thesis were modified as follows:

“TE is defined by Eq. (3) :”

Lines 399 to 405 of the thesis were modified as follows:

“In this paper, to obtain the total effect value of the space on the participants(henceforth abbreviated as TE), the Emotional Effect Value, Number of People Effect Value, Gender Structure Effect Value, and Age Structure Effect Value were normalized using SPSSAU 20.0 and combined. We utilized the total impact values for an overall assessment of the positive or negative impacts of different spaces on participants and the extent of the impacts, as well as for further research on the relationship between the impacts of spatial elements and participants.”

 

25. Line 370: Effects -> Emotional Effects? all people -> the crown / the participants

Response 25:

In response to this suggestion, lines 416 of the thesis were modified as follows:

“Emotional Effects of the spaces on the participants”

 

26. Line 379: What are “urban spacery” and “boats”? They are not included in the planar elements in 3.5.2.

Response 26:

In response to this suggestion, lines 425 of the thesis were modified as follows:

“…architecture, vessels…”

 

27. Line 381: this conclusion is not supported by figure 9. I suggest referring to figures 14 and 15, where you report the planar elements of different grids.

Response 27:

In response to this suggestion, lines 427 to 428 of the thesis were modified as follows:

“…(Figure 14 and 15).”

 

 

28. Line 383: not explained well what the EE value was calculated from the six scores in emotional analysis.

Response 28:

In response to this suggestion, lines 347 to 351 of the thesis were modified as follows:

“In our study, the EE value represented the emotional fluctuation data, which was derived from the likelihood data of all emotion categories outputted by the emotion recognition task. Emotion recognition in the paper involved categorizing a set of emotions and their likelihoods, and this output was used to calculate the EE value.

 

29. Line 385: Effects -> Emotional Effects? Is it still about the emotional effect in section 4.2 or total effects?

Response 29:

In response to this suggestion, lines 432 of the thesis were modified as follows:

“Total effects and emotional effects of the spaces on people of different ages

 

30. Line 397: Figure 10 has no names for x axis and has no y axis. The group names are not the same as the age groups described in the text. For discrete space units, I cannot see a reason for this curved graphic style. The white curve at the bottom is not understandable. It does not refer to any age group.

Response 30: The river diagram separated the different categories by color, and the corresponding values for each category were shown by the width of the "river". In this diagram, age groups were represented by different shades of blue, and the white curve were meaningless.

In response to this suggestion, lines 103 to 124 of the thesis were modified as follows:

“”

 

31. Line 389: group photo and graphic design elements are undefined.

Response 31:

In response to this suggestion, lines 436 to 439 of the thesis were modified as follows:

“…(“group photo” refered to a space where people were arranged in rows as if they were taking a photo in this study.) and street having a greater positive effect on children (Figure 11a). Wooden pavement with richer elements were more popular with children (Table 1).”

 

32. Line 394: what is “urban farscape”?

Response 32:

In response to this suggestion, lines 442 to 443 of the thesis were modified as follows:

“(“urban farscape” is defined as the natural or built-up landscape outside the site in this paper.),”

 

33. Line 395: “dotted tree” undefined.

Response 33:

In response to this suggestion, lines 444 to 446 of the thesis were modified as follows:

“Wooden decks with trees distributed in a dotted pattern in the space's floor plan were more popular with teenagers(Table 2).”

 

34. Line 399: Some elements remain undefined in the text such as “boys, girls, animals”. The numbers are not readable in the main text. You should either use a bigger font or split each sub figure into individual ones. I would also suggest considering moving these sub figures to supplementary material. Why the elements for correlation analysis are different for different age groups?

Response 34:

In response to this suggestion, lines 267 to 269 of the thesis were modified as follows:

“( In our study, these terms refered to different categories of spatial elements. Specifically, "boy" and "girl" refered to male and female children or adolescents, respectively, while "animal" refers to gulls at the research site.).”

In response to this suggestion, lines 448 of the thesis were modified as follows:

“”

 

35. Line 403: the title of table 1-5 are all the same. What are these tables? What does “distribution position” mean? How is the shape of the element acquired? The descriptions are all undefined in the text like “Faceted”, “Ribbon” etc.

Response 35:

In response to this suggestion, lines 453 of the thesis were modified as follows:

“Table 1. The state of distribution of the elements of the design plane of the urban space of children”

In response to this suggestion, lines 455 to 456 of the thesis were modified as follows:

“Table 2. The state of distribution of the elements of the design plane of the urban space of teenagers

In response to this suggestion, lines 473 to 474 of the thesis were modified as follows:

“Table 3. The state of distribution of the elements of the design plane of the urban space of young adults”

In response to this suggestion, lines 476 to 477 of the thesis were modified as follows:

“Table 4. The state of distribution of the elements of the design plane of the urban space of middle-aged adults”

In response to this suggestion, lines 487 to 488 of the thesis were modified as follows:

“Table 5. The state of distribution of the elements of the design plane of the urban space of the elderly”

In response to this suggestion, lines 328 to 333 of the thesis were modified as follows:

“Moreover, based on the results of spatial element identification, we summarized the distribution pattern characteristics of the elements, such as distribution in the form of points, lines, and surfaces (hereafter referred to as point-like distribution, ribbon distribution, and faceted distribution), as a variable for exploring the interaction between the elements and crowd preferences.”

 

 

36. Line 459: The phrases “causing” and “triggered” were used in this paragraph, which is not precise. This study, in my understanding, analyzed the correlation between variables in multiple aspects and emotional facial expressions. There is no further evidence to support casualties between them.

Response 36:

In response to this suggestion, lines 516 to 520 of the thesis were modified as follows:

“In the case of spaces where female participants experienced sadness, relatively more artificial  elements in the visual perception along with the absence of streetlights elements in the plane. In addition, it was found that hard surface in the visual perception accounted for more than 50% and there were no natural elements in the plane where female participants were fearful.

 

37. Line 460: “hard element” undefined.

Response 37:

In response to this suggestion, lines 525 to 530 of the thesis were modified as follows:

“(Note: In this paper, artificial elements were defined as man-made or human-involved elements, such as Group photo, House, Boys, Girls, Skyscraper, Plaza, Street Light Structures, Vending kiosks, Walking path, Urban Farscape, Street, Landmark, Seats, Vessels, Crowd, Park, Children, Classroom, Table, Pavilion, Bridge, Windows, etc.. Natural elements were defined as elements of nature, such as Trees, Sky, Water, Grassland, Plants, Cloud, Animals, etc., as in Figure 15.)

 

38. Line 465: the percentage of different planar elements are doubled in the top bar graphic and the middle table. In the table, these elements were called “plane element” while they were named planar elements in earlier text. Artificial and natural elements were not indicated in the bar graphics at the bottom.

Response 38:

In response to this suggestion, lines 523 of the thesis were modified as follows:

“”

Lines 536 of the thesis were modified as follows:

“

 

 

39. Line 472: It is not illustrated how the space units were selected for summarizing the male’s and female’s facial expressions. How did the different scenes lead to a fair comparison. Why not use the same space units?

Response 39: Thank you for your insightful comment concerning the selection of space units for summarizing male and female facial expressions. In our study, we aimed to understand gender preferences for space. To achieve this, we selected a variety of spaces categorized according to different moods, and analyzed facial expressions within these spaces to discern patterns of spatial preference or aversion for different genders. We understand your concern about the fairness of comparison when different scenes are used. Our intention was to explore a wide range of spatial types and moods to reveal how these factors influence gendered emotional responses. However, we agree that using the same space units for both genders could help to control for environmental variables and ensure a fairer comparison.In light of your feedback, we will add additional proof of the findings in future studies for different research purposes.

 

 

40. Line 501: diffe-rences -> difference

Response 40:

In response to this suggestion, lines 567 of the thesis were modified as follows:

“…difference…”

 

41. Line 501: When using the Chi-square test, the variables must be mutually exclusive. The spatial elements are apparently not exclusive in the scenarios. In this section, therefore, Chi-square test makes completely none-sense. In a very generous estimation, the author may aim at quantifying the effects of the design schemes. For this purpose, this study may use standard deviation of proportions of spatial elements against a target value acquired in this study. Even though, how these target values were calculated was poorly illustrated in this manuscript.

Response 41:

In response to this suggestion, Lines 27 to 29 of the thesis were modified as follows:

“The standard deviation of the difference results demonstrated that between the design op-tion(SD value=0.103) and the desired option for Space 1 were lower than current option(SD val-ue=0.129) and the expected scheme.”

Lines 562 to 572 of the thesis were modified as follows:

“To compare the attraction of spaces to young people before and after the design, The Non-parametric test and standard deviation of difference test were used to assess the value of the effect of spaces in this study. The nonparametric test of two independ-ent samples is a method of analyzing two sets of independent samples to infer wheth-er there is a significant difference between the distributions, etc., of the two aggregates from which the samples are drawn [71]. The most widely used measures of dispersion are the standard deviation (SD)[72,73]. In our study, the non-parametric test applied to calculate whether the pre- and post-design scenarios were characterized by signifi-cance with relation to the expected scenario. Subsequently, the difference standard deviation mean tests were utilized to test the difference between the before and after design scenarios and the expected scenario by using an online statistical tool (SPSSAU 23.0) [74].”

Lines 577 to 579 of the thesis were modified as follows:

“Table 6. The Non-parametric test and standard deviation of difference test for pre- and post-design space 1

Spatial Elements	Expect	Before	Diference		Expect	After	Diference
Sky	0.000	0.200	0.200		0.000	0.253	0.253
Landscape Structures	0.013	0.000	-0.013		0.013	0.107	0.093
House	0.184	0.215	0.030		0.184	0.173	-0.011
Trees	0.196	0.063	-0.133		0.196	0.240	0.044
Grassland	0.000	0.307	0.307		0.000	0.147	0.147
Road	0.061	0.215	0.154		0.061	0.053	-0.008
Water	0.162	0.000	-0.162		0.162	0.000	-0.162
Seats	0.104	0.000	-0.104		0.104	0.000	-0.104
Pavilion	0.104	0.000	-0.104		0.104	0.000	-0.104
Vessels	0.053	0.000	-0.053		0.053	0.000	-0.053
Brige	0.047	0.000	-0.047		0.047	0.027	-0.021
Kids	0.041	0.000	-0.041		0.041	0.000	-0.041
Plaza	0.018	0.000	-0.018		0.018	0.000	-0.018
Animals	0.015	0.000	-0.015		0.015	0.000	-0.015
P value			0.255		P value		0.441
SD value			0.129		SD value		0.103

“

Lines 583 to 585 of the thesis were modified as follows:

“Table 7. The Non-parametric test and standard deviation of difference test for pre- and post-design space 2

Spatial Elements	Expect	Before	Diference		Expect	After	Diference
Sky	0.225	0.324	0.099		0.225	0.303	0.078
Mountains	0.000	0.261	0.261		0.000	0.000	0.000
Trees	0.109	0.052	-0.057		0.109	0.099	-0.010
Water	0.205	0.363	0.158		0.205	0.158	-0.047
Cloud	0.000	0.000	0.000		0.000	0.092	0.092
Structures	0.122	0.000	-0.122		0.122	0.086	-0.037
House	0.102	0.000	-0.102		0.102	0.158	0.055
Seats	0.058	0.000	-0.058		0.058	0.000	-0.058
Pavilion	0.058	0.000	-0.058		0.058	0.105	0.047
Vessels	0.030	0.000	-0.030		0.030	0.000	-0.030
Brige	0.026	0.000	-0.026		0.026	0.000	-0.026
Kids	0.023	0.000	-0.023		0.023	0.000	-0.023
Plaza	0.021	0.000	-0.021		0.021	0.000	-0.021
Animals	0.020	0.000	-0.020		0.020	0.000	-0.020
P value			0.255		P value		0.427
SD value			0.101		SD value		0.046

 

The results of the nonparametric test indicates that the before(p=0.255) and after design(p=0.441) scenarios of space 1 and the before(p=0.255) and after design(p=0.427) scenarios of space 2 were significantly different from the expected scheme. As shown in Table 8, the standard deviation of the difference between the design option(SD val-ue=0.103) and the desired option for Space 1 were lower than its current option(SD value=0.129). Similarly, the standard deviation of the difference between the design option(SD value=0.046) and the desired option for Space 2 were lower than its present scenario(SD value=0.101). The results demonstrated that the design options for space 1 and space 2 had a higher likelihood of eliciting positive emotions of young people , implying that the reimagined urban environment garners a more affirmative recep-tion from the youthful demographic, thereby validating the efficacy of our design ap-proach.

Difference standard deviation test ….“

Lines 802 to 807 of the thesis were modified as follows:

“71. Beran, R.; Bilodeau, M.; de Micheaux, P.L.J.J.o.M.A. Nonparametric tests of independence between random vectors. 2007, 98, 1805-1824.

72. Livingston, E.H.J.J.o.S.R. The mean and standard deviation: what does it all mean? 2004, 119, 117-123.
73. Kao, L.S.; Green, C.E.J.J.o.S.R. Analysis of variance: is there a difference in means and what does it mean? 2008, 144, 158-170.
74. https://www.spssau.com, S.p.S.V.O.A.S.R.f.”

 

42. Line 513: “5”->”4”

Response 42: This section has been deleted due to adjustments in the validation methodology.

 

43. Line 519: “6”->”5”

Response 43: This section has been deleted due to adjustments in the validation methodology.

 

 

44. Line 549: Accuracy of the element recognition, facial recognition must be addressed in this section.

Response 44:

In response to this suggestion, lines 300 to 302 of the thesis were modified as follows:

“The methodology used in Tencent's Semantic Segmentation technology achieved state-of-the-art performance on challenging benchmarks, significantly outperforming previous methods with Mean Intersection over Union (MioU) up to 55.1% [25].”

Lines 374 to 375 of the thesis were modified as follows:

“Tencent face recognition model was an effective face recognition tool, achieving 99.80% accuracy at LFW 2017 [29].”

Lines 712 to 713 of the thesis were modified as follows:

“25. Jiang, Z.; Li, Y.; Yang, C.; Gao, P.; Wang, Y.; Tai, Y.; Wang, C. Prototypical contrast adaptation for domain adaptive semantic segmentation. In Proceedings of the European Conference on Computer Vision, 2022; pp. 36-54.”

Lines 720 to 722 of the thesis were modified as follows:

“29. Wang, H.; Wang, Y.; Zhou, Z.; Ji, X.; Gong, D.; Zhou, J.; Li, Z.; Liu, W. Cosface: Large margin cosine loss for deep face recognition. In Proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition, 2018; pp. 5265-5274.”

 

 

45. Line 580: delete “this study holds great significance and value for 580 both society and the design field, as”

Response 45:

In response to this suggestion, lines 638 to 639 of the thesis were modified as follows:

“Secondly, it empirically explored the relationship between urban space elements and residents' preferences and emotions.

 

46. Line 584: “external and internal pressures” undefined

Response 46:

In response to this suggestion, lines 639 to 641 of the thesis were modified as follows:

“The results indicate that a well-defined urban space design strategy can assist governments in attracting specific groups and establishing better communication with residents.”

Author Response File: Author Response.pdf

Reviewer 2 Report

The issues addressed in this article are highly original and in line with recent developments in deep learning and artificial intelligence. As far as we know, it is one of the first attempts to apply the outputs of the above-mentioned fields to the world of urban design, which is commendable in itself;

The research gap is well defined (151-152)

The analysis model is well constructed (165-173)

However, a few comments are addressed to the authors and require a response on their part, but a reading of the text reveals a general observation: despite the interest and rigour put into the research by the authors, the abusive use of superlatives cannot be overlooked. It would be more appropriate for the authors to rework the text and give it a more academic feel by reducing, if not eliminating, the abusive use of superlatives.

Here are a few examples:

“With continual iterative training, this model has exhibited exceptional capabilities that have surpassed human vision, particularly in the domain of object identification within images.”	116-118
“Furthermore, these studies did not take into account citizens' perceptions of utilizing small-scale public spaces from the perspective of the urban environment”	119-140	How can you be so sure? Have we looked at all the studies and what is there to prove it?
« Through extensive on-site investigations and observations, a wide range of space elements, spanning from the ethereal sky to the tangible plaza, have been meticulously curated and designated as the essential target elements for this spatial study.”	241-243
“Each of these elements is meticulously selected and targeted for a comprehensive study 255 on the principles of flat design for this space.”	255-256
“Artificial intelligence tools were harnessed in this study, …”	286	Is ENVI-met artificial intelligence-based?
Figure 2: Divided the Yangpu Binjiang public space into 10m*10m spatial units.	238	caption to be corrected
The diagrams in Figure 11 are not legible	399
“with classroom and water having a more significant positive effect on them”	410	????????????

 

 

The two p-values: Before (0.997467) and after (0.99968) are higher than 0.05 accepted value to reject the null hypothesis; how to explain that?

a few comments are addressed to the authors and require a response on their part, but a reading of the text reveals a general observation: despite the interest and rigour put into the research by the authors, the abusive use of superlatives cannot be overlooked. It would be more appropriate for the authors to rework the text and give it a more academic feel by reducing, if not eliminating, the abusive use of superlatives.

Here are a few examples:

“With continual iterative training, this model has exhibited exceptional capabilities that have surpassed human vision, particularly in the domain of object identification within images.”	116-118
“Furthermore, these studies did not take into account citizens' perceptions of utilizing small-scale public spaces from the perspective of the urban environment”	119-140	How can you be so sure? Have we looked at all the studies and what is there to prove it?
« Through extensive on-site investigations and observations, a wide range of space elements, spanning from the ethereal sky to the tangible plaza, have been meticulously curated and designated as the essential target elements for this spatial study.”	241-243
“Each of these elements is meticulously selected and targeted for a comprehensive study 255 on the principles of flat design for this space.”	255-256
“Artificial intelligence tools were harnessed in this study, …”	286	Is ENVI-met artificial intelligence-based?
Figure 2: Divided the Yangpu Binjiang public space into 10m*10m spatial units.	238	caption to be corrected
The diagrams in Figure 11 are not legible	399
“with classroom and water having a more significant positive effect on them”	410	????????????

 

 

Author Response

Response to Reviewer 2 Comments

 

Thank you for your professional insight into this paper and for your comments on revisions.

Based on your suggestions, we summarized the following 9 modifications:

 

1. 116-118 “With continual iterative training, this model has exhibited exceptional capabilities that have surpassed human vision, particularly in the domain of object identification within images.”

Response 1: This section has been deleted due to the restructuring of the methodological synthesis.

 

2. 119-140 “Furthermore, these studies did not take into account citizens' perceptions of utilizing small-scale public spaces from the perspective of the urban environment”How can you be so sure? Have we looked at all the studies and what is there to prove it?

Response 2:

In response to this suggestion, lines 146 to 147 of the thesis were modified as follows:

“Furthermore, most of these studies mentioned above did not take into account residents' perceptions of utilizing small-scale public spaces from the perspective of the urban envi-ronment [4].”

 

3. 241-243 “Through extensive on-site investigations and observations, a wide range of space elements, spanning from the ethereal sky to the tangible plaza, have been meticulously curated and designated as the essential target elements for this spatial study.”

Response 3:

In response to this suggestion, lines 260 to 262 of the thesis were modified as follows:

“Through extensive on-site investigations and observations, a wide range of space elements, spanning from the ethereal sky to the tangible plaza, have been curated and designated as the essential target elements for this spatial study.”

 

4. 255-256 “Each of these elements is meticulously selected and targeted for a comprehensive study 255 on the principles of flat design for this space.”

Response 4:

In response to this suggestion, lines 276 to 277 of the thesis were modified as follows:

“Each of these elements is selected and targeted for a comprehensive study on the principles of flat design for this space.

 

5. 286 “Artificial intelligence tools were harnessed in this study, …”Is ENVI-met artificial intelligence-based?

Response 5:

In response to this suggestion, lines 314 to 322 of the thesis were modified as follows:

“Supervised classification methods were considered to be relatively more precise in achieving quantitative categorization of remotely sensed data [60,61]. In this study,we applied supervised classification methods to classified the Google Earth sat-ellite images of the site. ENVI software 5.3 Region of Interest  (ROI) tools was employed for the comprehensive recognition and cluster analysis for supervised classifi-cation. We utilized the Support Vector Machine (SVM) algorithm within the ROI tool because it performed relatively better in extracting urban areas[62].

We created nine new ROI and selected multiple geometric sample ranges (geometry) for each land type ROI as training samples for supervised learning.”

 

6. 238 "Figure 2: Divided the Yangpu Binjiang public space into 10m*10m spatial units."caption to be corrected

Response 6:

In response to this suggestion, lines 257 of the thesis were modified as follows:

“Figure 2. 10m*10m spatial units of the site”

 

7. 399 The diagrams in Figure 11 are not legible

Response 7:

In response to this suggestion, lines 448 of the thesis were modified as follows:

“”

 

8. 410 "with classroom and water having a more significant positive effect on them”

Response 8: Thank you for your comment about the need for clarity regarding the term "classroom". In our study, we used the term "classroom" to describe outdoor landscaped spaces of a certain scale with stepped tables and chairs neatly arranged.

 

These outdoor classrooms are designed spaces that promote experiential learning and create a connection with the natural environment, impacting learners' engagement, well-being, and understanding of sustainability. From an urban planning perspective, these spaces can enhance landscape aesthetics, improve open space utilization, and contribute to the overall sustainability of an urban area.

In response to this suggestion, lines 461 to 463 of the thesis were modified as follows:

“(In our study, we used the term "classroom" to describe outdoor landscaped spaces of a certain scale with stepped tables and chairs neatly arranged, hereafter referred to as classrooms.)

 

9. The two p-values: Before (0.997467) and after (0.99968) are higher than 0.05 accepted value to reject the null hypothesis; how to explain that?

Response 9: Thank you for your expertise, but for accuracy reasons, we changed the chi-square test to the standard deviation of the difference, as detailed in the modifications below:

In response to this suggestion, Lines 27 to 29 of the thesis were modified as follows:

“The standard deviation of the difference results demonstrated that between the design op-tion(SD value=0.103) and the desired option for Space 1 were lower than current option(SD val-ue=0.129) and the expected scheme.”

Lines 562 to 572 of the thesis were modified as follows:

“To compare the attraction of spaces to young people before and after the design, The Non-parametric test and standard deviation of difference test were used to assess the value of the effect of spaces in this study. The nonparametric test of two independ-ent samples is a method of analyzing two sets of independent samples to infer wheth-er there is a significant difference between the distributions, etc., of the two aggregates from which the samples are drawn [71]. The most widely used measures of dispersion are the standard deviation (SD)[72,73]. In our study, the non-parametric test applied to calculate whether the pre- and post-design scenarios were characterized by signifi-cance with relation to the expected scenario. Subsequently, the difference standard deviation mean tests were utilized to test the difference between the before and after design scenarios and the expected scenario by using an online statistical tool (SPSSAU 23.0) [74].”

Lines 577 to 579 of the thesis were modified as follows:

“Table 6. The Non-parametric test and standard deviation of difference test for pre- and post-design space 1

Spatial Elements	Expect	Before	Diference		Expect	After	Diference
Sky	0.000	0.200	0.200		0.000	0.253	0.253
Landscape Structures	0.013	0.000	-0.013		0.013	0.107	0.093
House	0.184	0.215	0.030		0.184	0.173	-0.011
Trees	0.196	0.063	-0.133		0.196	0.240	0.044
Grassland	0.000	0.307	0.307		0.000	0.147	0.147
Road	0.061	0.215	0.154		0.061	0.053	-0.008
Water	0.162	0.000	-0.162		0.162	0.000	-0.162
Seats	0.104	0.000	-0.104		0.104	0.000	-0.104
Pavilion	0.104	0.000	-0.104		0.104	0.000	-0.104
Vessels	0.053	0.000	-0.053		0.053	0.000	-0.053
Brige	0.047	0.000	-0.047		0.047	0.027	-0.021
Kids	0.041	0.000	-0.041		0.041	0.000	-0.041
Plaza	0.018	0.000	-0.018		0.018	0.000	-0.018
Animals	0.015	0.000	-0.015		0.015	0.000	-0.015
P value			0.255		P value		0.441
SD value			0.129		SD value		0.103

“

Lines 583 to 585 of the thesis were modified as follows:

“Table 7. The Non-parametric test and standard deviation of difference test for pre- and post-design space 2

Spatial Elements	Expect	Before	Diference		Expect	After	Diference
Sky	0.225	0.324	0.099		0.225	0.303	0.078
Mountains	0.000	0.261	0.261		0.000	0.000	0.000
Trees	0.109	0.052	-0.057		0.109	0.099	-0.010
Water	0.205	0.363	0.158		0.205	0.158	-0.047
Cloud	0.000	0.000	0.000		0.000	0.092	0.092
Structures	0.122	0.000	-0.122		0.122	0.086	-0.037
House	0.102	0.000	-0.102		0.102	0.158	0.055
Seats	0.058	0.000	-0.058		0.058	0.000	-0.058
Pavilion	0.058	0.000	-0.058		0.058	0.105	0.047
Vessels	0.030	0.000	-0.030		0.030	0.000	-0.030
Brige	0.026	0.000	-0.026		0.026	0.000	-0.026
Kids	0.023	0.000	-0.023		0.023	0.000	-0.023
Plaza	0.021	0.000	-0.021		0.021	0.000	-0.021
Animals	0.020	0.000	-0.020		0.020	0.000	-0.020
P value			0.255		P value		0.427
SD value			0.101		SD value		0.046

 

The results of the nonparametric test indicates that the before(p=0.255) and after design(p=0.441) scenarios of space 1 and the before(p=0.255) and after design(p=0.427) scenarios of space 2 were significantly different from the expected scheme. As shown in Table 8, the standard deviation of the difference between the design option(SD val-ue=0.103) and the desired option for Space 1 were lower than its current option(SD value=0.129). Similarly, the standard deviation of the difference between the design option(SD value=0.046) and the desired option for Space 2 were lower than its present scenario(SD value=0.101). The results demonstrated that the design options for space 1 and space 2 had a higher likelihood of eliciting positive emotions of young people , implying that the reimagined urban environment garners a more affirmative recep-tion from the youthful demographic, thereby validating the efficacy of our design ap-proach.

Difference standard deviation test ….“

Lines 802 to 807 of the thesis were modified as follows:

“71. Beran, R.; Bilodeau, M.; de Micheaux, P.L.J.J.o.M.A. Nonparametric tests of independence between random vectors. 2007, 98, 1805-1824.

72. Livingston, E.H.J.J.o.S.R. The mean and standard deviation: what does it all mean? 2004, 119, 117-123.
73. Kao, L.S.; Green, C.E.J.J.o.S.R. Analysis of variance: is there a difference in means and what does it mean? 2008, 144, 158-170.
74. https://www.spssau.com, S.p.S.V.O.A.S.R.f.”

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thanks for all the efforts and improvements. I am impressed by the efficiency of the authors in addressing my comments. All my major concerns have been very well addressed. I am glad seeing this manuscript approaching the quality for publication.

The three objectives of this work, however, remain not well understood. These objectives are keys to effectively have an image of this work. Please find my comment below.

 

@Line 211-215

The research objectives are still not clear formulated. I would recommend the authors to read the examples in the link below and improve them again.

21 Research Objectives Examples (Copy and Paste) (helpfulprofessor.com)

 

@Line 247

Please check the slogan "One River, One River”?

Author Response

Response to Reviewer 1 Comments

Thank you for your professional insight into this paper and for your comments on revisions.

Based on your suggestions, we summarized the following 2 modifications:

 

1. Line 211-215

The research objectives are still not clear formulated. I would recommend the authors to read the examples in the link below and improve them again.

21 Research Objectives Examples (Copy and Paste) (helpfulprofessor.com)

Response 1:

In response to this suggestion, lines 211 to 215 of the thesis were modified as follows:

” Given the above, the objectives of this study were:

1) To construct a multi-dimensional research system of 10m*10m small-scale spatial units through combining field research and artificial intelligence analysis methods;

2) To examine emotional effects of space on age segmentation and gender segmentation based on analyzing data;

3) To evaluate the emotional effects of design solutions on specific groups of people based on the pattern of effects.”

 

2. Line 247

Please check the slogan "One River, One River”?

Response 2: Thank you for your comment regarding the slogan "One River, One River". The Shanghai Municipal Bureau of Planning and Natural Resources published the Area Construction Plan along the Huangpu River (2018-2035) and the Area Construction Plan along the Suzhou River (2018-2035) in 2021. According to the plan, this slogan was associated with the strategic development of the areas along the Shanghai Huangpu River and the Suzhou River Shanghai municipal section. It encapsulated the vision to integrate these areas, treating them as a single, cohesive unit for development purposes.

Author Response File: Author Response.docx