Dynamic Weights Based Risk Rule Generation Algorithm for Incremental Data of Customs Declarations
Round 1
Reviewer 1 Report (Previous Reviewer 2)
The manuscript has improved and comments and suggestions of this reviewer have been implemented. Readability and understanding of the proposed methods are much better now, hence, the manuscript can be accepted and printed in its current version.
Author Response
Thanks for taking time to review this article and reminding us the mistakes.
Author Response File: Author Response.docx
Reviewer 2 Report (Previous Reviewer 1)
The paper is revised according to the comments of the reviewer. It can be recommended for acceptance.
Author Response
Thanks for taking time to review this article and reminding us the mistakes. And we have made some changes based on the reviewers' comments.
Author Response File: Author Response.docx
Reviewer 3 Report (New Reviewer)
This paper proposed the Can-Tree incremental mining algorithm based on dynamic weights for the mining and generation of customs risk rules. I consider that the work is interesting, and the conclusions obtained have certain application potential.
1. It is suggested to change the title of the paper to "Dynamic weights based risk rule generation algorithm for incremental data of customs declarations".
2. I suggest the authors carefully read and improve the readability of this paper alongside the language.
3. The abstract is a litter noise and should be more straightforward for the reader regarding the proposed method and its motivation. The abstract should present some main points for the readers, such as the main contribution, the proposed method, the main problem, the obtained results, the comparative methods.
4. What is your main contribution? The contribution of the current work should be emphasized in the introduction.
5. Please check the correctness of Equation (1). This calculation formula is incorrect and non-standard.
6. Figure 2 are not clear, so they need to be redrawn and inserted with text in a graphic format.
7. The use of some mathematical formulas in this paper is not standardized, please correct them one by one.
8. The discussion of the results needs to include the strengths and weaknesses of the proposed algorithm.
Author Response
Response to Reviewer 3 Comments
This paper proposed the Can-Tree incremental mining algorithm based on dynamic weights for the mining and generation of customs risk rules. I consider that the work is interesting, and the conclusions obtained have certain application potential.
Point 1: Change the title
Response 1: Thanks for taking time to review this article and reminding us the mistakes. We change the title of the paper to "Dynamic weights based risk rule generation algorithm for incremental data of customs declarations".
Point 2: Rewrite abstract on page 1, line 10-23.
Response 2: We have reorganized the article abstract according to your comments (line 10-23).
Point 3: What is your main contribution? The contribution of the current work should be emphasized in the introduction.
Response 3: The main contributions are added in lines 119-130.
Point 4: Please check the correctness of Equation (1). This calculation formula is incorrect and non-standard.
Response 4: We have reworked Equation (1) as required (188).
Point 5: Figure 2 are not clear, so they need to be redrawn and inserted with text in a graphic format.
Response 5: We redrew Figure 2(194-200) and labeled each interval with a serial number in the figure.
Point 6: The use of some mathematical formulas in this paper is not standardized, please correct them one by one.
Response 6: We deliberate and make some modifications to Equation. 2-9 in lines 326-351.
Point 7: The discussion of the results needs to include the strengths and weaknesses of the proposed algorithm.
Response 7: We added some comments on the points and drawbacks of the algorithm in the experimental analysis section. (line 508-512, 533-536, 580-584)
Point 8: I suggest the authors carefully read and improve the readability of this paper alongside the language.
Response 8: We have revised the article for better readability.
Author Response File: Author Response.docx
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
In this work, the authors propose a risk-attribute combination weighted expansion method and an improved Can-Tree incremental mining algorithm (dynamic-weight Can-Tree) based on the sequential compressed storage of dynamic weights of data items.
The paper is well constructed. It has an adequate introduction, coupled with a presentation of the state-of-the-art in the field. The methodology section is extensive, and the results show the capabilities of the method proposed by the authors. I found this paper so interesting, it provides a useful contribution to its area of research.
However, a several details need to be corrected:
-acronyms need to be defined, for example: FOB, CIF please check the document
-figure 3 should be a flow chart not algorithm
-please consider the following: at the end of the introduction should briefly describe the sections of the document
-please reference the source of the data used in the document
-could the authors add a line in the document commenting on what are the limitation(s) of the methodology adopted in this work?
-check: 2. Materials and methods ??? (line 119-135)
-check: the result? (line 175)
-check: 3.3. dynamic-weight (line 339)
-check: further reduce The introduced (line 564)
-check (there is a replay): the introduced leaf node hash 563 table and incremental mining strategy further reduce
-there are many mistakes in the document, please check the document for possible spelling, grammar, and writing styles errors.
-check: complexity,2022,2022
please check and correct errors in the bibliography
Reviewer 2 Report
The manuscript deals with the problem of efficiently mining and generating customs risk rules that can handle the increasing number of customs declarations. Therefore, authors propose a Bi-KMeans-based approach in order to reduce memory footprint and provide better distinction of attributes.
Even the manuscript addresses an important and relevant topic, it needs extensive improvement for possible publication. In places the manuscript contains unfinished sentences (e.g. lines 22-25, 77-79 or 88-90), repetitions (looks like the final version was copied together from different parts from the authors) or parts that don't belong there (e.g. lines 119-135). Furthermore, the manuscript often contains very long sentences over several lines (e.g. lines 98-109) which makes reading and understanding very difficult. Authors are also *strongly* advised to consult a native speaker for an intensive language editing. In its current stage, the manuscript cannot be accepted for publication.
Concerning the proposed algorithms, a simple presentation in pseudo code would be easier to understand than a textual description. Furthermore, this reviewer wonders if authors also applied their algorithms on annotated data (i.e. were optimal clustering results were available) as any clustering by KMeans is a good as any other as long as the correct clustering is not know. Any comparison with other algorithms thus becomes irrelevant.