Combining Semantic and Structural Features for Reasoning on Patent Knowledge Graphs
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsI think it's a meaningful good paper.
Please revise and supplement the following matters.
1. It is necessary to add information on the use of patent data mining in the introduction.
2. It is not desirable to compare the improvement effect by changing the reference model. If the reference model is selected as Rotate among the four in Dataset 1 on page 14 (Table 2 and Figure 6), the same reference model should be compared in other datasets (page 15.16). It seems to be a trick to inflate the improvement effect.
And the efficiency should also be integer calculations. For example, 30% of the line 517 on page 14 should be modified to 27%, and 5% of the line 523 should be modified to 3%. Rounding up too randomly and making it big is the result of lowering the reliability of the paper.
The same reference model must be proportionally calculated for different datasets in the same way. Therefore, the ratio of the 14 page 532 line and the ratio of the 15 page 549 line must also be modified.
3. Figure 6 should be modified to Figure 7 in line 530 on page 14.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsComments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Comments and Suggestions for AuthorsDear Editor and Authors,
The paper "Combining Semantic and Structural Features for Reasoning on Patent Knowledge Graphs" proposes a method for analyzing patent knowledge graphs called SS-DSA. According to their paper, a patent graph can capture and display essential knowledge from documents, aiding in the verification of patent novelty and creativity.
The authors use three metrics—mean Rank (MR), Mean Reciprocal Rank (MRR), and Hits@n—to evaluate their proposal. [To authors] Why did you choose these metrics?
Line 44—Miric[1] et al. => I am not sure if this citation format meets the journal's standards. Please check the author information page.
From lines 44-58, the authors cite other authors' works. This part looks like the related works section. The same can be found in lines 75, 77, 82, 89, 94, etc.
There is nothing wrong with putting this information in the introduction. However, doing so makes a section on "Related Works" unnecessary. I suggest that the authors reorganize sections 1 and 2, separating the introduction from the "Related works", or else grouping them all into a single section.
Line 137. Add a period at the end of the figure caption (the same for all other figures).
Please remove the following lines that I believe were included incorrectly:
- Lines 195-197: "This section may be divided by subheadings. It should provide a concise and precise
description of the experimental results, their interpretation, as well as the experimental
conclusions that can be drawn."
Increase the size of Figure 2 (line 213). What do these colors mean? Explain in the caption.
Line 239, what does the symbol between V and R mean?
Figure 4 (line 292): why is the figure read from bottom to top? Is there a special reason for this?
"Section 4.1.1. Experimental Setup" (line 454) should not be part of the methodology section?
Table 1. Please align the numerical values to the right. This makes it easier to compare the values.
Line 494. Correct the text formatting.
Figures 6, 7, and 8 conflict with the adjacent tables' formatting.
Also, in Figures 6, 7, and 8: remove the borders of the graph. The secondary horizontal lines are unnecessary, but the main line of the x and y axes should be present. The size of the boxplots is quite tiny, so it is difficult to compare them. I suggest increasing the figure's size or the box's width.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 4 Report
Comments and Suggestions for AuthorsThe paper focuses on combining semantic and structural features for reasoning on patent Knowledge Graphs.
The topic is interesting and worth investigating. The introduction provides sufficient details about the topic. Sufficient technical details have been provided and the paper can be considered technically sound. However, several issues should be addressed:
- a paragraph could be added at the end of the introduction section, summarizing the paper.
- references should be provided for the models ComplEx, RotatE, MLMLM and DensE.
- at line 503, the metric "𝑛n" is mentioned, which does not seem to appear anywhere else in the paper. The authors are kindly asked to address the issue.
- four datasets have been used in the paper. The datasets should be made available using either a public repository or as supplementary material to the paper. Sharing the datasets is import for ensuring research reproducibility.
- additional details regarding the datasets and why they should be considered appropriate for validating the model should be included in the paper.
The paper includes sections that are probably left from the paper template and that should have been removed prior to the submission:
- "Research manuscripts reporting large datasets that are deposited in a publicly avail- able database should specify where the data have been deposited and provide the relevant accession numbers. If the accession num..."
- "The Materials and Methods should be described with sufficient details to allow oth- ers to replicate and build on the published results. Please note that the publication of your manuscript implicates that you must make all materials, data, computer code, and proto- cols associated with the pu..."
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsDear authors, thank you for responding to my questions. I noticed an improvement in the figures. However, the text in the captions sometimes mixes with the main text. I recommend adding a margin space to separate them.
I requested some changes in the introduction. However, I did not like the way the introduction was rebuilt. The text was quite long (4 pages). When I suggested merging the introduction and "related works" sections, this would imply adjusting the text and reducing content if necessary. Thus, I recommend separating these sections again if you want to keep the introduction with all this content. However, write a very succinct introduction, presenting the research problem and what will be addressed in the article (between 4 and 5 paragraphs). Then, present the "related works" section with the rest of the content.
The text has several formatting errors, especially the lack of use of spaces separating words. For example, see lines:
line 70 - Kambhatla[7]
line 223 - inference.We
line 227 - capabilities.Finally,
line 255-257 - the figure caption is blending into the text, which makes the result confusing
line 322 (and other figure captions) - remove the capital letter from the beginning of each word (except the first one).
line 577-578 - there is a changing in the font size. Please, correct this. Also, add a space between the figure caption and the main text.
line 601 - where is the caption for figure 7?
line 719 - include a section indicating where the data was obtained.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 4 Report
Comments and Suggestions for AuthorsI would like to thank the authors for the changes made.
Author Response
Dear reviewer,
On behalf of all the authors, I sincerely thank you for your letter and the reviewers' evaluation of our paper titled "Adaptive Attention Module for Image Recognition Systems in Autonomous Driving" (Manuscript ID: 3934270). We are deeply grateful for your recognition of our work. Your positive feedback and valuable suggestions have significantly contributed to the improvement of our manuscript.
Thank you very much!
Best wishes.
Your sincerely,
Kaitao Hu,
July 24 ,2024
School of Electrical and Electronic Engineering
Shanghai Institute of Technology
Shanghai 201418, China
19711277866
Round 3
Reviewer 3 Report
Comments and Suggestions for Authors
The article has received significant improvements. However, the "Data Availability Statement" section should contain links to locations where readers can download the data used in the research. This will ensure that the research is reproducible.
The authors only included the following statement: "The raw data supporting the conclusions of this article are available within the article."
In table 1, you describe four datasets. Where can I download them?
Dear Editor, I recommend accepting this paper if the authors include links to download the data used in the experiments.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
