Predictive Model with Machine Learning for Environmental Variables and PM_2.5 in Huachac, Junín, Perú

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors, here are recommendations for improvement of your paper:

- Repeated phrases  (e.g., "so machine learning algorithms can adequately process large amounts of data...").

- Incorrect PM2.5 instead PM_2.5, or NO2 instead NO₂...

Line 90-93 - Incorrect flow of the sentence: so machine learning algorithms can efficiently process large amounts of data and identify patterns and relationships between PM2.5 and meteorological variables [14]. so machine learning algorithms can adequately process large amounts of data

- Authors could elaborate how are they going to answer their research question. Authors could elaborate why this study is important for an international reader? Authors could elaborate their contribution to the literature and practice.

-Line 325-326 , 342-344, entire conclusion- Paragraph can not be made of only 1 sentence.

-Figure 10 is not visible.

-Research methodology is very modest and can be improved.

- Conclusion is very modest and can be improved.

 

 

Author Response

1. Summary

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted/in track changes in the re-submitted files.

 

Dear authors, here are some recommendations to improve your article:

- Repeated phrases (e.g. "so that machine learning algorithms can adequately process large amounts of data...").

- Incorrect PM2.5 instead of PM 2.5 , or NO2 instead of NO 2 ...

Corrected and replaced the correct words as indicated by the reviewer.

Line 90-93 - Incorrect sentence flow: so that machine learning algorithms can efficiently process large amounts of data and identify patterns and relationships between PM2.5 and meteorological variables [14]. so that machine learning algorithms can adequately process large amounts of data

Lines 90-93 were corrected to be coherent:

Weather stations and other environmental measurement devices generate a large amount of data that is difficult to process manually, and machine learning algorithms help process it efficiently to identify patterns and relationships between PM2.5 and meteorological variables

- Authors could explain how they will answer your research question. The authors could explain why this study is important for an international readership. The authors could explain its contribution to the literature and practice.

The wording of lines 103-106 was corrected as required.

Weather stations and other environmental measurement devices generate a large amount of data that is difficult to process manually, and machine learning algorithms help process it efficiently to identify patterns and relationships between PM2.5 and meteorological variables.

-Lines 325-326, 342-344, full conclusion- A paragraph cannot be made up of a single sentence.

The corrections were made which show lines 327-329 and include lines 344-346

-Figure 10 is not visible.

Figure 10 is made up of 4 plots (A, B, C, and D) and has been obtained by Rstudio.

-The research methodology is very modest and could be improved.

The methodology is designed and applied with the Lasso model (Least Absolute Shrinkage and Selection Operator) is a regression technique in machine learning and statistics that is used to perform both the selection of variables and the regularization of linear models whose advantages are simplicity and the handling of multicollinearity and its robustness with respect to linear regression. Other methodologies such as Ridge or Elastic Net do not consider the handling of multicollinearity drastically.

- The conclusion is very modest and could be improved.

It was written by improving the conclusions:

In conclusion, a predictive model with efficient machine learning was obtained for the prediction of PM2.5, which allows the analysis and measurement of the future contribution of decision making in the care of the environment.

 

Date of presentation

December 30, 2024

Date of this review

January 17, 2025 10:01:35

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors used a machine learning prediction model to simulate the AQI (Air Quality Index) in the Wachack-Huning area, considering the influence of multiple factors. However, the logic of the paper is unclear. I recommend that the authors reorganize the structure and improve the clarity of the language.

 

Comments and suggestions on the authors are as follows:

1.      Is the format of the abstract a specific requirement of the journal? If not, it is suggested to avoid the detailed listing of methods and results in the abstract. Focus instead on highlighting the novelty and contributions of the method, and briefly mention the key findings. Additionally, the language should be reorganized to ensure logical coherence and avoid an overly conversational tone.

2.      Could you explain why the first citation in the paper starts from [2] instead of [1]?

3.      It is suggested that the introduction focuses on the research problem you are addressing. Specifically, what is the problem, what method have you proposed, and how does it differ from previous studies? There is no need to repeatedly emphasize the harm caused by pollutants, as this does not add value to the paper.

4.      If the regression method used in the study is not a novel approach, there is no need to dedicate a large portion of the paper to describing its principles. If there are any improvements to the method, they should be emphasized in the discussion.

5.      The second section describes many theoretical methods, but I do not see any innovative points, nor do I understand the main focus of this section. At the very least, the authors should clearly describe the logic and process of the methods they plan to use.

6.      The figures should be presented more clearly, especially by using better legends and explanations. In addition, the output results shown in Figure 9 should be better organized and explained, rather than just providing screenshots of the software output interface, which is unprofessional.

 

7.      The overall structure of the paper should be clearer, particularly with regard to the quantitative and qualitative analysis of the results. At a minimum, the readers should be able to easily extract the main conclusions of the study.

Comments on the Quality of English Language

It is suggested that the language be polished and the presentation's logistics improved.

Author Response

3. Point-by-point response to Comments and Suggestions for Authors

The authors' comments and suggestions are as follows:

1. Is the format of the abstract a specific requirement of the journal? If not, it is suggested to avoid detailed enumeration of methods and results in the abstract. Instead, the focus should be on highlighting the novelty and contributions of the method, and briefly mentioning the key findings. In addition, the language should be reorganized to ensure logical coherence and avoid an overly conversational tone.

Corrections in the summary were addressed by removing the numbering of the parts

It was also corrected on line 46.

2. Could you explain why the first citation in the article starts with [2] instead of [1]?

Corrected the enumeration sequence from 1.

3. It is suggested that the introduction focus on the research problem being addressed. Specifically, what is the problem, what method has been proposed, and how is it different from previous studies? There is no need to repeatedly emphasize the harm caused by contaminants, as this does not add value to the article.

The problem is air pollution by particulate matter over the environment and this article aims to predict the behaviors of PM2.5 particulate matter, applying machine learning methodology in order to make decisions for the benefit of the surrounding population.

4. If the regression method used in the study is not a novel approach, it is not necessary to spend a large part of the article describing its principles. If there are improvements in the method, they should be highlighted in the discussion.

The LASSO (least absolute selection and contraction operator) and Ridge regression have been applied, as shown in the results lines 372 and 388. Also the Elastic Net model regression in line 407, as predictive models.

5. The second section describes many theoretical methods, but I do not see any innovative point nor do I understand the main objective of this section. At the very least, the authors should clearly describe the logic and process of the methods they intend to use.

In this research, it was found that the lasso (RMSE=25.36206), Ridge (RMSE=25.36471) and Elastic net (RMSE=25.36203) models are efficient predictors and the best model found is lasso with an α=0.1111111 and a Lambda value λ=0.150025 is represented in equation 28.

6. The figures should be presented more clearly, especially by using better legends and explanations. In addition, the output results shown in Figure 9 should be better organized and explained, instead of simply providing screenshots of the software output interface, which is unprofessional.

Figure 9 has been removed, we are placing the results from the R Studio program

> summary(lm)

Call:

lm(formula = .outcome ~ ., data = dat)

Residuals:

    Min        1Q       Median      3Q      Max 

-53.087  -19.664   -5.102    16.336  177.352 

Coefficients:

                      Estimate   Std. Error   t value     Pr(>|t|)     

(Intercept)     83.40932    1.07213    77.798   < 2e-16 ***

Humidity       -10.37024    0.23201  -44.698   < 2e-16 ***

Temperature   -0.12717    0.03169   -4.013   6.04e-05 ***

---

Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 25.37 on 13269 degrees of freedom

Multiple R-squared:  0.1478,      Adjusted R-squared:  0.1477 

F-statistic: 1151 on 2 and 13269 DF, p-value: < 2.2e-16

In these results, the significance of the regressor variables was highly significant with a p-value less than2×〖10〗^(-16) allowing us to propose the multiple regression model for PM2.5 shown in equation 25.

7. The overall structure of the work should be clearer, particularly regarding the quantitative and qualitative analysis of the results. At the very least, readers should be able to easily draw the main conclusions of the study.

We appreciate your feedback and have reviewed and corrected the relevant points.

 

Comments on the quality of the English language

It is suggested that the language be polished and the logistics of the presentation improved.

Date of submission

30 December 2024

Date of this review

17 January 2025 03:14:47

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Please check the attachment.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

To some extent, some sentences are difficult to understand, see review.

Author Response

Overall, readers may not be clear about the author's research purpose. Therefore, the author needs to carefully consider and clarify the research purpose of this article again. If the research objective is to compare models, the author should fully demonstrate the superiority of the proposed model compared to other models. In the introduction section, the author should delve into the content of the relevant models, point out the shortcomings of the current model, and emphasize the necessity of this study.

 

In Huachac, Junín, air pollution, especially PM 2.5 particulate matter, represents a significant health and environmental risk. However, there is limited capacity to pre-dict and manage this pollutant due to the lack of accurate predictive models that inte-grate relevant environmental variables. This hinders the implementation of effective pollution control and mitigation policies. Therefore, it is necessary to develop a predic-tive model based on Machine Learning techniques to anticipate PM 2.5 levels and op-timize environmental management in the region.

The problem is atmospheric pollution by particulate matter on the environment and this article aims to predict the behavior of PM2.5 particulate matter, applying ma-chine learning methodology in order to make decisions for the benefit of the sur-rounding population.

The study provides a predictive machine learning model of PM2.5 which contributes to the predictive understanding of the behavior of environmental variables in the study areas, which will allow international bodies to use it as a model for analysis

 

If the research objective is to apply the proposed model to predict environmental variables and PM2.5 concentration in specific regions, the author should provide specific prediction results. Meanwhile, the discussion and conclusion sections should be modified accordingly based on the research objectives. From the title, it seems that the author has established a prediction model for environmental variables and PM2.5 concentration in the region. However, from the content of the article, although the model has been established, no prediction results have been provided, indicating a lack of application examples of the model.

 

The best predictive model found is quite robust given that it meets the indicators in favor of the model LASSO regression analysis with a high AIC and BIC and a low RMSE for the 16,543 data analyzed. This is the thirteenth model emerged as the best predictor for PM2.5 Air Quality Index (AQI) with specific parameters: Alpha: 0.1111111 and Lambda: 0.150025. The model predicts a decrease in PM2.5 AQI by 10.3022000 units per year, for temperature changes, PM2.5 AQI decreases by 0.12688124 units, the Mean Squared Error (MSE) for training is 25.36255, while for testing it is 25.84308, resulting in a mean difference of 0.48053, this indicates an im-pressive prediction efficiency of 98.1406% when comparing training to testing out-comes. These results highlight the effectiveness of the regression models in predicting PM2.5 AQI and provide a basis for further analysis and refinement of predictive tech-niques in environmental monitoring.

 

 

In the discussion section, the author seems to want to discuss the performance of the model, but has not clearly demonstrated its superiority over other models, thus failing to highlight the characteristics and value of the research. The article should focus more on presenting knowledge or innovative points that can move readers.

 

Based on existing knowledge, the model that typically demonstrates the best effi-ciency in predicting PM2.5 AQI is often determined through comparative analysis of various regression techniques, such as Ridge, Lasso, and Elastic Net regression, and the Key Factors Influencing Model Efficiency are[43].

1. Model Selection: The choice between Ridge, Lasso, and Elastic Net can impact performance based on the nature of the dataset.
2. Penalty Values: Adjusting penalty values can optimize each model's ability to prevent overfitting or underfitting.
3. Feature Selection: Models like Lasso are particularly effective at feature selec-tion, which can enhance predictive accuracy.
4. Cross-Validation: Using techniques like k-fold cross-validation helps in as-sessing model performance more reliably.

In fact, to determine the best-performing model for PM2.5 AQI prediction, one would typically analyze metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared values across different models and configurations. The model yielding the lowest error metrics and highest R-squared value would be considered the most efficient for this specific prediction task. [42], [44], [45], [46].

 

 

In addition, the article seems to only describe the patterns of a phenomenon, and it is difficult to see the value and significance of the research from the abstract, nor does it highlight the characteristics of the research. The abstract only provides model related parameters, but the ultimate goal of modeling is application, so the abstract should include research conclusions or application results.

 

 

The best predictive model found is quite robust given that it meets the indicators in favor of the model LASSO regression analysis with a high AIC and BIC and a low RMSE for the 16,543 data analyzed. This is the thirteenth model emerged as the best predictor for PM2.5 Air Quality Index (AQI) with specific parameters: Alpha: 0.1111111 and Lambda: 0.150025. The model predicts a decrease in PM2.5 AQI by 10.3022000 units per year, for temperature changes, PM2.5 AQI decreases by 0.12688124 units, the Mean Squared Error (MSE) for training is 25.36255, while for testing it is 25.84308, resulting in a mean difference of 0.48053, this indicates an im-pressive prediction efficiency of 98.1406% when comparing training to testing out-comes. These results highlight the effectiveness of the regression models in predicting PM2.5 AQI and provide a basis for further analysis and refinement of predictive techniques in environmental monitoring.

In conclusion, a predictive model with efficient machine learning was obtained for the prediction of PM_2.5, which allows the analysis and measurement of the future contribution of decision making in the care of the environment.

 

 

 

In the introduction, the author failed to clearly articulate the profound significance of this study. The article does not explicitly state the source of the data.

 

 

The study was conducted in the district of Huachac, province of Chupaca, department of Junín - Peru, where data were collected from the HUAYAO monitoring station using geolocation tools and software with Python libraries with a sample of 16,543 records that were conditioned and normalized with the EDA metolodology.

 

The accuracy of the English expression in the article needs to be further improved, such as repeated descriptions in lines 91-94.

The wording and formatting have been corrected.

 

 

For the abbreviation "EDA" in line 138, if it is not an abbreviation of the preceding noun, then this writing style may not be appropriate. The author should clearly explain its meaning to avoid confusion for readers.

 

Exploratory Data Analysis (EDA), allows a clear view of the data, performing de-scriptive analysis, adjustment of variable types, detection and treatment of missing data, identification of outliers and correlation of variables [22]. In descriptive analysis, one dives into the data to understand its structure and nature. This includes generating summary statistics, such as mean, median, standard deviation, and creating graphs that reveal the distribution of the data. This is followed by adjusting the types of varia-bles in which the data can be presented in different formats and types (numerical, cat-egorical, dates, etc.) [23]. In this phase, the variables are adjusted and transformed as necessary to ensure that they can be used effectively in the analysis. In addition, miss-ing data is detected and processed, where incomplete data can be a challenge. During EDA, we identified observations with missing data and considered options such as imputation or deletion of rows or columns to ensure completeness of results, as well as identified outliers, known as outliers. The EDA helped us to identify and understand these values to determine if they should be treated or if they are legitimate and signifi-cant. Finally, correlation of variables was done to explore the relationships between variables, using statistical techniques and visualizations to discover connections and patterns that can reveal valuable information about the particulate matter problem analyzed in the research [24], [13].

 

 

 

 

Corresponding explanations should be provided for Figures 10, 11, 12, 13, etc., explaining their significance and function. Meanwhile, the quality of the images also needs to be improved to ensure that readers can understand them clearly.

 

The resolution of the images and the corresponding explanation have been corrected and increased.

 

Finally, the selection of references should be more up-to-date and avoid using outdated literature to reflect the cutting-edge and applicability of the research.

 

The references have been updated in recent years.

 

In summary, the author needs to further revise and improve the article to clarify the research objectives, highlight the research features, improve the accuracy of English expression, and

 

Provide specific prediction results and conclusions. At the same time, attention should also be paid to improving the quality of charts and the timeliness of references.

 

 

All your comments have been answered as requested.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This paper is a research report on air pollution, mainly discussing the status quo and impact of air pollution and the use of machine learning model to predict PM2.5 concentration. The paper points out that air pollution is a global problem that has a serious impact on human health and ecosystems, especially in urban and rural areas. The machine learning algorithm is also used to process a large amount of data, analyze the relationship between humidity and temperature and PM2.5 air quality index (PM2.5AQI), and establish a prediction model. This study has certain academic value and provides important theoretical support and practical guidance for promoting environmental protection and sustainable development, but there is still room for further modification and improvement.

 

1) Please delete blank lines such as 44,52,61,66,73,79, and do not appear between sections.

2) It is recommended that the "keyword" font be bold, and the following keywords are sorted according to the first letter.

3) It is recommended to carefully check the "Materials and Methods" section, repeat the 2.2 title, and delete the space before the 2.3 title and a number of basic standard issues.

4) Figure 1 in the text has an obvious grain feeling, and some words are not clear. It is recommended to re-insert the more clear picture and carefully check whether the other pictures need to be changed.

5) It is suggested to synthesize the original 2.2 "Research Methods" section in the original 2.5 "Exploratory Data Analysis" section.

6) The original 2.2 "Machine Learning Model" part and the original 2.3 part are uniformly summarized as "research methods", and add the three-level subtitle.

7) Please remove the extra number of 223 lines and remove 224 lines.

8) It is suggested that when describing the results, quantitative data analysis should be provided as much as possible, such as how the specific values are obtained.

9) The discussion and conclusion section suggests that the shortcomings and limitations of the supplementary article should make it more scientific and reasonable.

10) The innovative nature of this study is not sufficient and it is recommended that additional innovative elements of the study be elaborated.

11) It is suggested that the format of the references in the text is unified, do not appear in the sentence and some outside the sentence.

12) Some of the years of the references are not bolded, please check them carefully and revise them.

Author Response

This paper is a research report on air pollution, mainly discussing the status quo and impact of air pollution and the use of machine learning model to predict PM2.5 concentration. The paper points out that air pollution is a global problem that has a serious impact on human health and ecosystems, especially in urban and rural areas. The machine learning algorithm is also used to process a large amount of data, analyze the relationship between humidity and temperature and PM2.5 air quality index (PM2.5AQI), and establish a prediction model. This study has certain academic value and provides important theoretical support and practical guidance for promoting environmental protection and sustainable development, but there is still room for further modification and improvement.

 

• Please delete blank lines such as 44,52,61,66,73,79, and do not appear between sections.

Lines 44, 52, 61, 66, 73, 79 and other blank lines of the article were eliminated.

• It is recommended that the "keyword" font be bold, and the following keywords are sorted according to the first letter.

The wording was made according to the requirements of the format.

• It is recommended to carefully check the "Materials and Methods" section, repeat the 2.2 title, and delete the space before the 2.3 title and a number of basic standard issues.

The free spaces were eliminated and part 2.2 was restructured in accordance with point 6 of your comments.

 

• Figure 1 in the text has an obvious grain feeling, and some words are not clear. It is recommended to re-insert the more clear picture and carefully check whether the other pictures need to be changed.

Figure 1 has been replaced by one with a higher resolution.:

• It is suggested to synthesize the original 2.2 "Research Methods" section in the original 2.5 "Exploratory Data Analysis" section.

Exploratory Data Analysis (EDA) provides a comprehensive understanding of data by performing descriptive analysis, adjusting variable types, handling missing data, identifying outliers, and analyzing variable correlations [38]. Descriptive analysis involves summarizing data with statistics like mean, median, and standard deviation, and visualizing distributions. Variables are then adjusted or transformed (e.g., numerical, categorical, dates) to suit analysis needs [39]. Missing data is detected and addressed through methods such as imputation or row/column deletion to ensure reliable results. Outliers are identified to assess whether they require treatment or hold significant meaning. Lastly, variable correlations are analyzed using statistical methods and visualizations to uncover patterns and relationships, offering valuable insights into the particulate matter problem explored in the study [40].

2.2.2. Machine Learning Models

Machine learning, a branch of AI, develops algorithms and models to enable com-puters to learn and make data-driven decisions without explicit programming, fostering autonomous and efficient machine learning. [17]. ML and statistics are closely related, with ML relying on many statistical principles and methods.[18]

According to Aliaj [24] LASSO (Least Absolute Shrinkage and Selection Operator) regression is a regularization technique for variable selection and model complexity reduction in machine learning and statistics. It's an extension of linear regression with a penalty to avoid overfitting [25]. LASSO is used when there are many features, helping in variable selection while fitting a predictive model [26]. The LASSO loss function combines the linear regression loss (mean square error) and the absolute value of feature coefficients multiplied by a hyperparameter λ [27]

• The original 2.2 "Machine Learning Model" part and the original 2.3 part are uniformly summarized as "research methods", and add the three-level subtitle.

The parts have been restructured and synthesized as follows:

2.2. Research methods

2.2.1. Exploratory EDA data analysis

2.2.2. Machine learning models (lasso, Ridge and Elastic net).

2.2.3. Metrics for validation of predictive models.

• Please remove the extra number of 223 lines and remove 224 lines.

Additional lines were eliminated.

• It is suggested that when describing the results, quantitative data analysis should be provided as much as possible, such as how the specific values are obtained.

According to the results of the algorithms applied in the processing of the data, the various metrics are obtained and the parameters of each model applied are presented in the tables.

• The discussion and conclusion section suggests that the shortcomings and limitations of the supplementary article should make it more scientific and reasonable.

Conclusions and discussions were increased:

In summary, the choice of penalty values in regression models is crucial for achieving optimal performance by managing the trade-off between bias and variance, influencing both model complexity and predictive power.

The best predictive model found is quite robust given that it meets the indicators in favor of the model LASSO regression analysis with a high AIC and BIC and a low RMSE for the 16,543 data analyzed. This is the thirteenth model emerged as the best predictor for PM2.5 Air Quality Index (AQI) with specific parameters: Alpha: 0.1111111 and Lambda: 0.150025. The model predicts a decrease in PM2.5 AQI by 10.3022000 units per year, for temperature changes, PM2.5 AQI decreases by 0.12688124 units, the Mean Squared Error (MSE) for training is 25.36255, while for testing it is 25.84308, resulting in a mean difference of 0.48053, this indicates an impressive prediction efficiency of 98.1406% when comparing training to testing outcomes. These results highlight the effectiveness of the regression models in predicting PM2.5 AQI and provide a basis for further analysis and refinement of predictive techniques in environmental monitoring.

• The innovative nature of this study is not sufficient and it is recommended that additional innovative elements of the study be elaborated.

They were added in the methodology:

Based on existing knowledge, the model that typically demonstrates the best effi-ciency in predicting PM2.5 AQI is often determined through comparative analysis of var-ious regression techniques, such as Ridge, Lasso, and Elastic Net regression, and the Key Factors Influencing Model Efficiency are

1. Model Selection: The choice between Ridge, Lasso, and Elastic Net can impact per-formance based on the nature of the dataset.
2. Penalty Values: Adjusting penalty values can optimize each model's ability to pre-vent overfitting or underfitting.
3. Feature Selection: Models like Lasso are particularly effective at feature selection, which can enhance predictive accuracy.
4. Cross-Validation: Using techniques like k-fold cross-validation helps in assessing model performance more reliably.

In fact, to determine the best-performing model for PM2.5 AQI prediction, one would typically analyze metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared values across different models and configurations. The model yielding the lowest error metrics and highest R-squared value would be considered the most efficient for this specific prediction task.

• It is suggested that the format of the references in the text is unified, do not appear in the sentence and some outside the sentence.

The format was revised according to the requirements.

• Some of the years of the references are not bolded, please check them carefully and revise them.

Years in bold type have been revised according to the following format.

 

Submission Date

30 December 2024

Date of this review

16 Jan 2025 07:40:05

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The paper is improved and acceptable for publication in this form.

Author Response

The paper is improved and acceptable for publication in this form.

We appreciate the review and acceptance of our article.

Thank you

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The suggestions for the previous version have been addressed. I think it can be accepted.

Author Response

3. Comments and Suggestions for Authors

The suggestions for the previous version have been addressed. I think it can be accepted.

We appreciate the review and acceptance of our article.

Thank you

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

check the revieThe author made some efforts to make changes, and the quality of the article has been improved.

Author Response

Comments and Suggestions for Authors

check the review The author made some efforts to make changes, and the quality of the article has been improved.

Your suggestions and comments have been considered and the wording and structure have been improved. We appreciate the review and decision on our article.

Thank you

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This paper is a research report on air pollution, mainly discussing the status quo and impact of air pollution and the use of machine learning model to predict PM2.5 concentration. The paper points out that air pollution is a global problem that has a serious impact on human health and ecosystems, especially in urban and rural areas. The machine learning algorithm is also used to process a large amount of data, analyze the relationship between humidity and temperature and PM2.5 air quality index (PM2.5AQI), and establish a prediction model. This study has certain academic value and provides important theoretical support and practical guidance for promoting environmental protection and sustainable development, but there is still room for further modification and improvement.

 

1) As long as the word ‘keywords’ is bolded, followed by the keywords in accordance with the first letter of the alphabet can be, do not need to add additional bolding.

2) It is recommended that the original section 2.2.1 and Figure 2 be consolidated in section 2.3 to describe the analysis.

3) It is suggested to standardise the format of the mathematical formulas in the second part of the Materials and Methods to make the article neater and more aesthetically pleasing.

4) It is suggested that the serial number 3.3 before the results section in line 218 should simply be changed to 3.

5) Delete the original line 219 and proceed directly to the descriptive analysis of the results.

6) Present the formatting of the charts in Part III consistently, e.g. Figure 3 is cited topically, while Figure 4 is cited centrally, and so on to double-check other details.

7) It is recommended that the limitations of this study be more clearly stated in the conclusions to provide food for thought for subsequent research.

8) Use the formatting brush to make changes to the references in strict accordance with the journal format and remove the blank line in 649.

Author Response

This paper is a research report on air pollution, mainly discussing the status quo and impact of air pollution and the use of machine learning model to predict PM2.5 concentration. The paper points out that air pollution is a global problem that has a serious impact on human health and ecosystems, especially in urban and rural areas. The machine learning algorithm is also used to process a large amount of data, analyze the relationship between humidity and temperature and PM2.5 air quality index (PM2.5AQI), and establish a prediction model. This study has certain academic value and provides important theoretical support and practical guidance for promoting environmental protection and sustainable development, but there is still room for further modification and improvement.

 

• As long as the word ‘keywords’ is bolded, followed by the keywords in accordance with the first letter of the alphabet can be, do not need to add additional bolding.

Fixed additional bold text.

• It is recommended that the original section 2.2.1 and Figure 2 be consolidated in section 2.3 to describe the analysis.

It was corrected according to the suggestion indicated.

• It is suggested to standardise the format of the mathematical formulas in the second part of the Materials and Methods to make the article neater and more aesthetically pleasing.

 

The format of mathematical formulas was standardized in the second part of Materials and Methods.

• It is suggested that the serial number 3.3 before the results section in line 218 should simply be changed to 3.

 

The indicated numbering was corrected and changed.

• Delete the original line 219 and proceed directly to the descriptive analysis of the results.

 

The line was eliminated

• Present the formatting of the charts in Part III consistently, e.g. Figure 3 is cited topically, while Figure 4 is cited centrally, and so on to double-check other details.

This was corrected based on comments, suggestions and considering the formats established in the journal template.

• It is recommended that the limitations of this study be more clearly stated in the conclusions to provide food for thought for subsequent research.

According to the monitoring protocol of the Ministry of Environment of Peru (MINAM), the number of monitoring stations must be proportional to the size of the population. To overcome this limitation, a greater number of monitoring stations should be considered in future research.

8) Use the formatting brush to make changes to the references in strict accordance with the journal format and remove the blank line in 649.

It has been corrected according to the suggestion and in accordance with the format of the journal and line 249 has been removed.

Submission Date

30 December 2024

Date of this review

01 Feb 2025 12:30:25

Author Response File: Author Response.pdf