Review paper from Journal of Network and Computer Application
28 views
Skip to first unread message
Muhammad Royyan Zahir
Oct 9, 2017, 2:28:34 AM10/9/17
to nsl...@googlegroups.com
Dear colleagues,
--
I have already received review and decision regarding my journal after few months. Unfortunately, it was rejected. Hope you guys could learn from my mistakes. With this, it attached the review.
Reviewers' comments:
The referees made the following points about this paper.
Reviewer #1: Thanks to authors for addressing the comments. However, i still have the following guidelines for the authors
1. A separate comprehensive literature review section must be in the paper. According to the revised version author just explained only three related works [13,14,15]. Further, it is hard to believe that the authors only found three papers in this area as related work.
2. The English of the paper is still not convincing. The authors should consider some native English speaker to improve the English of the paper. For example, the sentence in the revised version "Fault detection system is a monitoring system that identifies when a fault has occurred and pinpointing its location" can be written as "Fault detection system is a monitoring system that identifies a fault and its location when it occurred". An other sentence in revised version "In system accuracy, there is two type of simulation". Kindly correct these.
3. More justification in terms of number of related papers should be inserted into the paper that such kind of work is only possible to evaluate via simulation instead of publicly available data-set.
Reviewer #2: In this paper, the authors proposed a scheme to detect faulty node for industrial network. The design was based on Markov Chain in which interval between two consecutive errors was used to determine a node's state. The current version has the following issues, which makes it unsuitable for publication in this journal:
1. Interval between two consecutive errors is a well-known indicator that has been widely used in faulty node detection. In this work, the authors introduced three Markov states to describe this indicator. It is not fundamentally different from the traditional methods. Thus, I do not think this work has a remarkable innovation.
2. Some experiment results were shown in the paper. However, the authors did not give the deatails on experimental environment, experimental conditions and configuration. Thus, it is difficult to prove the effectiveness of the proposed scheme.
3. For the Markov model, the authors did not show how to determine the parameters of the model. For some constants, like P_{0,0}, the authors didi not give the reasons for them.
4. This paper is hard to read. The organization and presentation need to be greatly improved.
The referees made the following points about this paper.
Reviewer #1: Thanks to authors for addressing the comments. However, i still have the following guidelines for the authors
1. A separate comprehensive literature review section must be in the paper. According to the revised version author just explained only three related works [13,14,15]. Further, it is hard to believe that the authors only found three papers in this area as related work.
2. The English of the paper is still not convincing. The authors should consider some native English speaker to improve the English of the paper. For example, the sentence in the revised version "Fault detection system is a monitoring system that identifies when a fault has occurred and pinpointing its location" can be written as "Fault detection system is a monitoring system that identifies a fault and its location when it occurred". An other sentence in revised version "In system accuracy, there is two type of simulation". Kindly correct these.
3. More justification in terms of number of related papers should be inserted into the paper that such kind of work is only possible to evaluate via simulation instead of publicly available data-set.
Reviewer #2: In this paper, the authors proposed a scheme to detect faulty node for industrial network. The design was based on Markov Chain in which interval between two consecutive errors was used to determine a node's state. The current version has the following issues, which makes it unsuitable for publication in this journal:
1. Interval between two consecutive errors is a well-known indicator that has been widely used in faulty node detection. In this work, the authors introduced three Markov states to describe this indicator. It is not fundamentally different from the traditional methods. Thus, I do not think this work has a remarkable innovation.
2. Some experiment results were shown in the paper. However, the authors did not give the deatails on experimental environment, experimental conditions and configuration. Thus, it is difficult to prove the effectiveness of the proposed scheme.
3. For the Markov model, the authors did not show how to determine the parameters of the model. For some constants, like P_{0,0}, the authors didi not give the reasons for them.
4. This paper is hard to read. The organization and presentation need to be greatly improved.
--
Muhammad Royyan
Networked Systems Lab., NSL: http://nsl.kumoh.ac.kr
School of Electronic Eng. and Dept. of IT Convergence,
Kumoh National Inst. of Tech., S. Korea
Reply all
Reply to author
Forward
0 new messages