All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
The combination of fuzzy logic and machine learning could help to diagnose diseases with overlapping symptoms.
No more comments
No more comments
No more comments
No more comments
The manuscript needs a stronger introduction, improved visuals, and a more detailed methodology. It should address overfitting of results, and expand conclusions (please, also add some statement about limitations, and strategies for coping with them in the future).
**PeerJ Staff Note:** Please ensure that all review and editorial comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** The review process has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
I have the following minor observations or queries and comments which may further enhance your piece of work. The authors require to modify the following points in detail.
1. In abstract, please bring in your 2-3 special quantitative achievements from the results of this study in context of environment by mixing up the research objectives and problems. Please make your abstract within 250 words only. Also, check spellings for many words, which are wrong and are written in hurry.
2. The introduction part is required to add few more sentences to increase the strength of this article and kindly bring in the research problem, objective, novelty and explain it in last paragraph of the section of Introduction.
3. Add few more sentences in the very beginning of introduction explaining about your paper's contribution or attempts for dealing or presenting solutions for a specific problem/s and your special contribution with this research paper.
4. Please present the methodology section in a compact graphical format as well.
5. Literature review part is very weak kindly revise it.
6. Please present your literature review in a concise SmartArt chart format.
7. Just after the Methodology, please mention the benefits of your research for the society at large with respect to evaluating its key determinant.
8. The section of "Results" must explain research problems, solutions and the contribution of your study theoretically with around 500-750 words.
9. Please insert graphical presentations for your results.
10. Explain why have you deployed this study in a separate section of "Policy suggestions" just before the section of "Conclusions".
11. Please add serial numbers to the headings of sub sections. and focus on discussion of 'MFFCG–Multi feature fusion for hyperspectral image classification using graph attention network' and 'Deep Learning with Graph Convolutional Networks: An Overview and Latest Applications in Computational Intelligence' as new methods of deep learning
12. Add three more paragraphs (at least of 250 words excluding the existing one) in the section of conclusions mentioning the limitations of the study and remedies to limitations with achieved objectives after conducting this study.
13. Add around 150 words more to the section of conclusions explaining the future scope of your research study, limitations if any faced while conducting your research and the procedure to remove the limitations of research.
14. Also, explain all the tables more briefly and the explanations of each table in the sections of "Results".
15. Use of English language is very poor. Revise almost all sentences in the manuscript with appropriate use of grammar, punctuation and speech (active or passive voices).
16. Please do not use multiple nouns and verbs in a single sentence for putting more pressure to damage the real and simple meaning of your sentences.
As above
As above
As above
1. Please changes the title. I make suggestion: Fuzzy Evaluation and Explainable Machine Learning for Diagnosing Rheumatic and Autoimmune Diseases.
2. In abstract add clear statements about methods, dataset size, and novelty.
3. Add one sentence on the practical impact of explainable AI (XAI) in clinical settings.
4. I suggest a full English proofreading pass, ideally with help from a professional editor or Grammarly.
5. Provide more context on how the dataset is balanced post-ADASYN.
6. Discuss potential biases due to dataset origin (Iraqi hospitals) and generalizability.
7. Explain why you chose 12 models. Were they benchmarked based on literature? Provide some proofs.
8. Provide a concrete example of how fuzzy ratings from experts translate into rankings.
9. Add captions that summarize the key takeaway of each figure, not just "ROC for models."
10. PCA Figure 4 needs better explanation: How much variance is captured by PC1 and PC2?
11. Table 1 could benefit from standardizing units more clearly and separating continuous and categorical features.
This manuscript presents a novel approach combining explainable machine learning models with a fuzzy evaluation (FDOSM) framework to improve the diagnosis of rheumatic and autoimmune diseases using tabular clinical data. The research is timely and relevant, particularly given the growing interest in interpretable AI for medical applications. The manuscript's strengths include its comprehensive evaluation across multiple machine learning models and metrics, as well as its emphasis on explainability.
no comment
no comment
However, several areas need improvement:
1. The use of explainable AI (XAI) is a major strength. However, the discussion around XAI lacks depth—particularly regarding how the global and local interpretations from the selected model (GBoost) influenced or could influence real-world clinical decisions.
2. While the dataset seems extensive and well-labeled, more detailed information is needed regarding the exact preprocessing steps. The description of ADASYN and handling of imbalanced classes is helpful, but the implementation lacks reproducibility due to missing code references and detailed data split rationale.
3. There are several grammatical and syntactic issues throughout, particularly in the abstract and introduction. For example, “alongside with fuzzy assessment framework” should be revised to “alongside a fuzzy assessment framework.”
4. The manuscript presents a strong evaluation framework using multiple metrics (e.g., F1, MCC, Kappa). However, the overfitting analysis—while detailed—raises concern as many models, including top-performing ones like GBoost and XGBoost, show high training accuracy with a notable validation gap.
5. Figures such as the PCA plot and PR curves are useful, but some (e.g., Figures 10 & 11) could benefit from better resolution and labeling. Also, Figure captions need to be more informative—some simply state what is being shown without interpreting its relevance.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.