A Novel Komodo Mlipir Algorithm and Its Application in PM2.5 Detection
Round 1
Reviewer 1 Report
This paper investigated a novel Komodo Mlipir algorithm and its application in PM 2 2.5 detection. This paper is interesting and the method is novel, however there exist some issues the authors should clarify in the coming version.
1. Some comparison experiments should be added to show the advantages over the existing ones.
2. Some remarks should be added to show the advantages and disadvantages of the proposed algorithm.
3. Some works about RL should be added to improve the quality of the literature review such as Multiobjective reinforcement learning-based neural architecture search for efficient portrait parsing; Neural architecture search for portrait parsing.
4. There are some typos in this paper such as below equations (2) and (4), "Where" should be "where".
Author Response
Dear Editors and Reviewers:
Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “A Novel Komodo Mlipir Algorithm and Its Application in PM 2.5 Detection”. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our research. We have studied comments carefully and have made correction which we hope to meet with approval. Revised portions are marked in red in the paper. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Responds to the reviewer’s comments:
Reviewer #1
Point #1 Some comparison experiments should be added to show the advantages over the existing ones.
Answer:Taking into account the reviewer's opinion, we added the random forest prediction algorithm in the traditional machine learning algorithm to predict PM2.5 in the actual application of PM2.5 prediction. The accuracy of the actual result is still 1.34% worse than the algorithm in this paper.
Point #2 Some remarks should be added to show the advantages and disadvantages of the proposed algorithm.
Answer:thanks for valuable comments, in section 1, we add “ KMA can balance exploration and exploitation effectively, can have faster convergence speed, and can accurately converge to the optimal value by using adaptive population. However, KMA still has the problem of local convergence in the solution of complex functions, such as converging to a local optimum in the Rastrigin function.”
Point #3 Some works about RL should be added to improve the quality of the literature review such as Multiobjective reinforcement learning-based neural architecture search for efficient portrait parsing; Neural architecture search for portrait parsing
Answer:Thanks to the reviewers for their valuable suggestions, we added five references to improve the quality of the references. In Section 4.1 and Section 5, respectively, newly added literature 25 and literatures 30-32,35.
Point #4 There are some typos in this paper such as below equations (2) and (4), "Where" should be "where".
Answer:Dear reviewer, I am very sorry, there is a typo in the article, we have changed 'Where' to 'where'. And we've re-read the full text, checking for spelling, grammar, and more.
Author Response File: Author Response.docx
Reviewer 2 Report
The manuscript is well drafted and the result section is well explained with graphs, tables and literature.
Figure 2 explains the overall methodology of the work
Table 1-3 needs detailed elaboration on why to use these functions and their purpose with applications
Table 4-8, I really dont get it. Why this data is presented in tabular form, it will be more significant if the comparative analysis is provided in form of graphs as shown in figure 4-6
Author Response
Dear Editors and Reviewers:
Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “A Novel Komodo Mlipir Algorithm and Its Application in PM 2.5 Detection”. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our research. We have studied comments carefully and have made correction which we hope to meet with approval. Revised portions are marked in red in the paper. The main corrections in the paper and the responds to the reviewer’s comments are as flowing:
Responds to the reviewer’s comments:
Reviewer #2
Point #1 Table 1-3 needs detailed elaboration on why to use these functions and their purpose with applications
Answer:Thanks for your valuable comments, we have added detailed explanations of some benchmark functions in Table 1-3, and cited literature with detailed explanations of benchmark functions. The effect of some benchmark functions are explained. in section 4.1, we add “ The benchmark function is a set of functions to test the performance of the evolutionary computing algorithm. The benchmark functions are described in detail by Yao Xin et al.” and “ In addition to the global minimum, the F1 function also has d dimension local minimum values, which are continuous, concave, and unimodal. The F9 function, called the Rastrigin function, has numerous local minima. In two-dimensional form, the F10 function is characterized by an extremely flat outer area and a large hole in the center. The introduction of the F10 function puts the optimization algorithm at risk of getting stuck in numerous local minima. The F11 function, called the Griewank function, has numerous local minima that are regularly distributed. The rescaled form of the F18 function has a mean of zero and a variance of one, and also adds a tiny Gaussian error term to the output.”
Point #2 Table 4-8, I really dont get it. Why this data is presented in tabular form, it will be more significant if the comparative analysis is provided in form of graphs as shown in figure 4-6
Answer:We are very sorry for the negligence in the presentation. We have adjusted Table 4-8, marked the best mean and standard deviation with bold and underlined, highlighting the superiority of the algorithm, especially Table 8, in order to intuitively display the algorithm in terms of superiority in high dimensions, we express the curve convergence of the benchmark function under 1000 dimensions in the form of a graph.
Author Response File: Author Response.docx