Assessment of Investment Attractiveness in European Countries by Artificial Neural Networks: What Competences are Needed to Make a Decision on Collective Well-Being?

Round 1

Reviewer 1 Report

In this paper, the authors build an artificial intelligence model to assess the foreign direct investment attractiveness of some European countries and the competences needed to manage this kind of models.

Introduction:

The introduction lacks focus on the problem analyzed and in some cases, there are generalizations regards the scientific literature without citing that literature (i.e. line 47).

There is an apparent contradiction in the research question formulated with two different proposals (lines 71 and 74).

There are some sentences difficult to understand as the final part of the introduction (line 87)

Literature review:

In the literature review there are some seemingly unrelated references to the research questions as paragraphs one, two and three, starting at line 93.

There are some repetitions of the facts explained in the introduction.

The authors explain that one of the limitations of the use of artificial intelligence is the lack of data, when that problem is the same for any statistical technique.

There are references to ‘hard and soft skills’ without explaining their meaning.

There are some sentences that are not related with the research question (sentences starting at line 168 and line 187)

Research methodology:

The authors explain that they selected some factors without explaining why.

The authors state that they use 2 methods for hierarchical classification but explain the use of more than 2 methods.

The data used is not described, not even the sample size, which makes very difficult to replicate the analysis.

Discussion:

The conclusions are not clear nor linked to the research question.

There are two ‘Table 4’ and the second one is not referenced in the manuscript.

There are two different sets of limitations not interconnected between them.

Conclusions:

The authors do not explain why the selected competencies are required. (line 467)

Minor issue:

The article of: Snieska, Burksaitiene, & Zykiene (2019) is not in the reference.

Author Response

We are sending the corrections according to the reviewers suggestions. 

Author Response File: Author Response.pdf

Reviewer 2 Report

I read the article with interest, and I appreciated it. However, I believe that there is a substantial problem. The article, as is presented, is well structured for the part concerning the "literary review", but, in my point view, lacks for the part related to the "operational research" performed by the authors.

In par. 4 "Research methodology" from what I could understand, the authors start from a research by one of the authors in collaboration with others [Snieska, Burksaitiene, & Zykiene (2019)]; it is not clear whether the “Factors and indicators determining country investment investment attractiveness to foreign direct investment "have been elaborated by the authors in this new article, or are taken back “tout court” by the previous mentioned research; this aspect needs to be clarified.

The main problem, however, is that in par. 4 a series of things are declared, but there is no evidence.

A scientific article, to be an article and not a simple literary review, must make a progress, of a methodological type, to something already existing and tested. In your case I seem to understand that there is a model, and that it is also an evolution of previous works: you don’t demonstrate, however, how the model proposed works, the differences with the starting models; if necessary it is also appropriate to give appropriate specifications under the "formal-mathematical" profile in order to better understand how the proposed model/method works.

In this sense, it is not possible for the reader to verify the work performed. The reader, in this sense, must "trust” the authors.

I believe that you have to solve this lack.

I therefore suggest the authors to expand the par. 4 giving more evidence of the constructed and used model, of how it works and how the data have been processed and how you have reach the results.

With this upgrade, this work can become a scientific article; otherwise it can be published but as is as "literary review", because anyway it is a good work.

Author Response

We are sending the corrections according to the reviewers suggestions. 

Author Response File: Author Response.pdf

Reviewer 3 Report

It is not clear the criteria used to group the european countries (Table 2).

Besides, the fact of putting together countries inside and outside the European Union could distort the results as nations with different attractiveness are compared (think of the possible European Union funding)

Author Response

We are sending the corrections according to the reviewers suggestions. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear Authors, first of I congratulate you for the improvement of your manuscript, nevertheless there are some issues that could be addressed.
Languaje:
I understand that the paper has been proofread and in that sense there is no problem, but I recommend the authors to rewrite some sentences that are difficult to understand, i.e, in line 37:
'The inquiry shows that competences under the smartness approach are making new connections between artificial intelligence and economic analysis'
or line 61-62:
'The development of investment attractiveness by standard factors already is not enough'

Abstract:
I recommend to summarise the abstract with the key points of the research and move some concepts to the introduction.
The abstract uses the term 'smartness' but it is not defined, in my case I dont understand the concept.

Section 3:
Its a little bit confusing the mix of concepts from line 243 on, talking about cities or regions, when the goal of the research is to analyse countries.

Section 4:
You talk about factors and main components but the technique used to create the factors is not specified.

Section 5:
'Such use of artificial neural networks would result in a larger sample of training; therefore, artificial neural networks could accurately record relationships among individual indicators, but these artificial neural networks would require much longer training' I dont see a problem with the training time if we improve the results.

Section 6:
'A faster and more accurate answer, compared with the currently used methods' Which methods?

Author Response

The corrections are done according to the comments

Author Response File: Author Response.pdf

Reviewer 2 Report

The text has been improved. A little more indepth explanation of the "mathematical structure" of the methodological approach could be useful (even if not essential).

Author Response

The corrections are done according to the comments

Author Response File: Author Response.pdf

Round 3

Reviewer 1 Report

Dear authors, congratulations for the improvements on your manuscript, for me the article is fine in its current form.