Hybrid Decision Models of Leasing Business for Thailand Using Neural Network

Round 1

Reviewer 1 Report

The article is very interesting in terms of the purpose it pursues, however modeling decision making of leasing business based on personal aspects is complicated from neural networks. The article is presented is very dense and certain images should be in an appendix since they take up a lot of space and the representativeness is not relevant since they are already explained. 

Abstract: It does not really indicate what is intended to be achieved in the research, but refers directly to the experimentation as such.

Background: ATM acronyms are not defined, there are spelling mistakes and incorrect grammatical expressions. This section is too long and includes aspects that should be in an introduction section. It is difficult to read, being necessary to segment it. Literature review is extensive.

Research Methodology: I consider this section to be correct in three phases. Perhaps a comparison could be made by varying the percentage of the training and test set.

After all the specifications and all the tests that have been performed, the results obtained should be compared with previous works, since being a work that consists of different parts, results can be compared separately. This comparison should be in the discussion section, indicating that the results are better or worse and what may be the reason, which in part, is due to the lack of sufficient data to train the neural network.

A conclusion section is needed.

Author Response

I have already send your comment in attachment

Author Response File: Author Response.docx

Reviewer 2 Report

In this paper, a dynamic model of leasing business is established to improve the effectiveness of leasing business decision-making. The article is informative, but I have the following questions and suggestions for the paper:

(1)  Questionnaire survey was used many times in the article, but the content of questionnaire survey was not put into the appendix.

(2)  In the approval model, the data used in the questionnaire survey is a little old, so it is recommended to provide the data in recent years.

(3)  In the research methods, SWOT analysis, PEST analysis and five forces analysis are required to study the relationship between preference factors and recognition factors, but these three analysis methods are not reflected in the paper.

(4)  Some figures in the paper are not clear enough, such as "Figure 7 system dynamics cost model".

(5)  In "3.1.3-3.1.4 Process", inconsistent spacing between paragraphs.

(6)  Both approval model and preference model have model accuracy and error value test, but hybrid decision model lacks this part.

(7)  Some terms are not explained in the article, such as "K-cross validation", "EPOC" and "DEA-DA".

Author Response

i have already sent your comment in attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Please highlight the changes made to the manuscript. 

Author Response

Comment and suggestion ffie in attachment

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

The authors have followed the recommendations and improvements have been made to the manuscript.
In my opinion the manuscript can be published, however there are several errors. For example, Figures 2 and 6 are not completely visible.

Please revise the manuscript in its entirety and correct these errors.

Author Response

comment and suggestion in attachment fille

Author Response File: Author Response.docx