Short Words for Writer Identification Using Neural Networks
Round 1
Reviewer 1 Report
Abstract: Abstract needs to be improved following problem under study, research methods, results, conclusion and significance.
Formatting and style for paper writing can be improved. Methodology and Instrumentation should be included.
Abbreviations should be defined at first mention and used consistently thereafter. Such as NIST, CNN etc.
NIST Special Database 19 (SD19) was used in the study but there is not citation or acknowledgement or approval stated.
Contribution of the article should be mentioned clearly.
Conclusion: The conclusion needs to be rewritten mapping the objectives, limitations and future work.
As there are works done for English, Chinese and Arabic, authors are doing it for English or other language. Difficult to identify the novelty. It is stated in the conclusion “handwriting information which is irrelevant of the language used.” Arabic used (right to left) so the essence will be lost.
References should be used from resent years and the need for extensive literature review is required so that the comparison form other works can be reflected in the article. There are only 3 articles referencing from 2021-2022. There are few articles suggested that might be of interest.
https://doi.org/10.3390/app10227999
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In the present work writer verification is carried out based on short length words and neural networks. The author is inspired the topic by the need that exists several times to recognize/distinguish a person only from one of his words that he may have left in a small note. The paper is well presented.
In letter-based alphabet writing, this method is very simples; it may be used to identify “a kind of the people” rather then “a person”. For example, in the school, the pupils are trained to write the letter following almost the teacher’s manner. But for character-based character writing, such as Chinese, the proposed method may be interesting, but there are other problems. Anyway, the proposed solution is interesting for discuss.
There are no evaluation results by using ANN to test the proposed method. For example, in the abstract: “It is proved that the neural methods 19 achieve identification performance which approaches 100%”. But in the paper, there is no strong prove by the result to support this conclusion.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The introduction part might be expanded to include difficulties that arise in the context of previous work related to the words that are used in the prediction of actions.
The use of issues in the existing system on the various neural networks and how the author proposes to overcome them should be included in the literature review approaches.
The independence of the method from the language used is based on the fact that the 125 features to be extracted reveal the orientation of the lines along the word that make up the 126 letters of the word, it would be better the authors can address with a proper justification.
Authors should show a clear representation of how the proposed algorithm is efficient in handling the issues compared to the previous algorithms.
Authors should justify why they have considered only NIST Database and its features.
The scope of future work should be stated in the conclusion.
Authors may refer the below papers for having a better literature review.
IoT enabled HELMET to safeguard the health of mine workers
Evaluation of efficacy and safety of subcutaneous acetaminophen in geriatrics and palliative care (APAPSUBQ)
Discuss future intentions in light of the current state of research and its constraints.
Authors might look to some recent high-quality works in the literature review.
A few figures seem to be blurred and the authors should take care of the clarity.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I have carefully reviewed the changes made in response to the previous comments. However, I regret to inform you that I find the revised manuscript to be unsatisfactory due to the lack of adherence to the comments and the minimal effort put into addressing the issues raised.
Firstly, I would like to reiterate the importance of updating the references with recent publications from 2023. It is crucial for a scholarly manuscript to reflect the current state of the field and to incorporate the latest research findings. Despite my previous suggestion, I see that you have not made any substantial updates to the references. This omission significantly weakens the validity and reliability of your research, as it fails to integrate the most relevant and up-to-date sources.
Additionally, I am disappointed by the minimal effort put into addressing the concerns raised in the previous review. Many of the issues highlighted in the previous comments have only been superficially modified.
To ensure the scholarly merit of your manuscript, I strongly urge you to reconsider the reviewer comments and devote sufficient effort to revising the manuscript accordingly. It is essential to update the references with recent publications to demonstrate your familiarity with the current literature. Furthermore, I request that you carefully address all the concerns raised in the previous review, providing detailed explanations or making necessary changes to the methodology, analysis, and discussion as appropriate.
Should you choose to submit a revised version, please provide a point-by-point response to the previous review, clearly indicating the changes made. This will aid in the evaluation of your revisions and facilitate further decision-making.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
