IDEINFO: An Improved Vector-Weighted Optimization Algorithm
Round 1
Reviewer 1 Report
Please see the attached file.
Comments for author File: Comments.pdf
Author Response
Thank you for suggestions, please see attached file.
Author Response File: Author Response.doc
Reviewer 2 Report
In this study, authors proposed an improved INFO (weighted-mean-of-vectors) algorithm to solve non-convex and highly non-linear problems that are given in the literature. They incorporated three different techniques to improve this stochastic algorithm: a two-stage backward learning strategy to enhance the search capability of IDEINFO, a differential evolution strategy with a greedy mechanism to improve local search, and a combination of t-distribution and probabilistic strategy to avoid entrapment in local optima. 14 benchmark test examples were considered in the experimental study to test the proposed algorithm. The IDEINFO proved its accuracy and robustness.
Assuming that the INFO algorithm was discovered in 2022, this topic is up-to-date and deserves appreciation. However, there are several points I would like authors to consider before the final acceptance of the manuscript. Here are my suggestions:
- Many words in the text are joined, making the paper slightly difficult to read. Also, additional spacing between equations is desired to improve readability.
- State-of-the-art review of the literature should be expanded by adding more recently published works. Here are some relevant examples that can be included:
https://doi.org/10.3390/app13020945
https://doi.org/10.3390/s22103810
https://doi.org/10.1109/ACCESS.2022.3153493
- What is INFO1 and INFO2? They are used in the experimental study but not previously defined. It is mentioned in the Conclusions that they represent two variants of the improved INFO algorithm. Authors should consider the possibility of defining what algorithms are used in the experimental study when testing the performances of the IDEINFO. INFO1 and INFO2 are the focus of the study and, therefore, should be clearly specified prior to experimental results.
- Discussion is perhaps the weakest part of the manuscript. 14 test functions were considered in the experimental study, and the obtained results are summarized in three tables regarding high-latitude SO functions, high-latitude MO functions, and low-latitude benchmark functions, respectively. The authors should elaborate more on statistical measures and numerical values in these tables rather than just stating that INFO1 and INFO2 showed superior accuracy and robustness.
- The limitations and future research directions should be more elaborated in the Discussion part or before the Conclusions.
Author Response
Thank you for suggestions, please see attached file.
Author Response File: Author Response.doc
Round 2
Reviewer 1 Report
The paper is written very well, results are discussed in detailed, work is presented in smooth way which will attract readers from different labs. The paper is revised according to suggestions and concerns. No more comments for changes from my side, paper in current situation seems acceptable for publication.