Review Reports

Round 1

Reviewer 1 Report

The paper is well structured and it apparently contributes with exciting results. However, the most essential part of the paper is the label definition of the medical recommendation, based on the EHS, which is not clearly explained. Therefore is hard to measure the effectiveness of this manuscript. It is not clear, what kind of outputs are defined in the model, and how the ML algorithm reaches those.  Therefore, I tend to recommend rejection, unless that part is well and clearly fixed 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper presented an improved approach to have better recommendations in the medication domain which should also consider the case of Drug-Drug-Interactions(DDI). The key idea about the improvement is the use of ontology-like tree structure with embedding approach on the graph. In the evaluation, the authors used the dataset that were openly available and often used in other researches in this field. The given resuls shown in Table3, Table4, Figure6, Figure7 and Table5, it is clear that the proposed approach achieved better results to the baseline approaches (i.e., [21],[22],[44],and [45]).   One thing the reviewer would note is, the author argued the tree-structure used in the work is "ontlogy tree". This wording is quite misleading since they are not actually used as "ontology" which standard meaning is a meta-model to construct models and actually it seems an embedding on a knowledge graph. In this paper, there is no clear explanation about why it should be said as "ontology" tree. Even no references about what is an ontology and what is an "ontology tree" in this paper. The reviewer recommend the author not use the term "ontology tree" but use "knowledge graph" instead, and change the title accordingly.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

 

The paper presents a system for medication recommendation, based on the use of ontologies tree and learning algorithms.

 

The paper presents cleanly in sections 1 and 2 the problem and the related work. In section tree, the model proposed in explained in depth. Section 4 provides a detailed analysis of the performance of the system when compared with others introduced in previous sections. The experiments done are explained and also the methods and datasets used in the evaluation. The performance of the proposed system is better in general terms than the others, providing in this way a justification for the new system.

 

In general terms, the paper clearly presents the proposal. Nevertheless, Conclusion sections are  short and may be improved by adding more detail in what are the contributions of the paper and the proposal. Moreover, it is important to introduce what are the drawbacks of the system and the future work. 

 

Finally, a general revision of the English language  must be done. Although, the paper is easy to follow and understandable, there are some grammatical errors that can be improved.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

the authors addressed the issues found in the reviews

Author Response

Thanks again for your valuable advice.