Spotting anomalous trades in NFT markets: The case of NBA Topshot

Konstantinos Pelechrinis; Xin Liu; Prashant Krishnamurthy; Amy Babay

doi:10.1371/journal.pone.0287262

Peer Review History

Original SubmissionNovember 15, 2022
6 Feb 2023 Decision Letter - Junhuan Zhang, Editor PONE-D-22-31512Spotting Anomalous Trades in NFT Markets: The Case of NBA TopshotPLOS ONE Dear Dr. Pelechrinis, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Mar 23 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Junhuan Zhang, PhD Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Partly Reviewer #2: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: No Reviewer #2: I Don't Know ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ****** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: The research of NFT is currently in a relatively new and popular situation. This paper studies the NFT market and its transactions from the perspective of abnormal transaction recognition, which is of great significance. However, in terms of the format and content of the paper, there are still some key points that need to be further improved, so as to support this paper to become a more solid and standardized academic paper. In terms of the format of the paper, the contents to be improved include the following: 1. In the last paragraph of the first part, the hyperlink references to the content of the following chapters are displayed incorrectly and need to be modified. 2. In addition, the captions and pictures of the paper can be placed together rather than separately. In addition, the clarity of each picture in the paper should also be improved. 3. The content in the subtitles of each section of the paper needs to be enriched with corresponding descriptive text, rather than just one or two sentences. For example, "Circulation Count", "Player", "Bought Price" and so on in Section 2.2.1 need to be enriched. More importantly, the content of the article needs to be modified, including the following points 1. The data used can be updated. This paper applies the transaction data from 07/27/2020 to 03/19/2021. You can consider applying the latest transaction data to this paper. 2. The application of the conclusion is limited. According to the research conclusion of the paper, it is not possible to accurately evaluate whether a transaction is abnormal. 3. The paper does not consider the user's transaction logic and transaction behavior, but only considers the static transaction price and other information. You can also consider building the behavior model of transaction users/agents. 4. In the result analysis part, the probability threshold (1%) selected in the paper lacks the necessary theoretical support and detailed description. 5. It can be considered further about the underlying blockchain technology corresponding to NFT, such as smart contracts, to enrich the judgment method and conclusion interpretation of abnormal transactions. There are still quite a few problems in this paper, but the research topic of this paper is of significance. After further revision, I hope this paper can become a good academic paper. Reviewer #2: The paper is written in a clear and concise manner, with proper technical terms used where necessary and no grammatical errors. The objective and methodology are clearly stated and easy to understand. The topic of NFT markets and their potential for illegal activities is highly relevant and timely, and the focus on the NBA TopShot marketplace is unique and appropriate. The methodology used to build the framework for labeling transactions as anomalous or not is well-explained and appears to be appropriate for the problem at hand. The results of the network analysis are well presented and provide evidence for the authors' conclusions. However, it would be helpful to have a further explanation of the network metrics used and how they support the findings. The abstract mentions that the transactions labeled as anomalous will need to be further audited, so it would be helpful to discuss potential avenues for future work. ****** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: Yes: Wenbin Wu ******** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0287262.r001
Revision 1
26 Apr 2023 Author Response * Response is also provided as a pdf for better formatting and the addition of a figure that cannot be added in the text response ** Reviewer #1 The research of NFT is currently in a relatively new and popular situation. This paper studies the NFT market and its transactions from the perspective of abnormal transaction recognition, which is of great significance. However, in terms of the format and content of the paper, there are still some key points that need to be further improved, so as to support this paper to become a more solid and standardized academic paper. In terms of the format of the paper, the contents to be improved include the following: 1. In the last paragraph of the first part, the hyperlink references to the content of the following chapters are displayed incorrectly and need to be modified. We have fixed the hype refences in the revised manuscript. 2. In addition, the captions and pictures of the paper can be placed together rather than separately. In addition, the clarity of each picture in the paper should also be improved. We would like to thank the reviewer for this comment. The separation of the captions and images was according to the journal’s submission guidelines. In fact, we attempted to included them inline with the text in the revised version but the editorial office returned the manuscript with the request to change this. This would have helped to showcase that the quality of the images was simply distorted because they were zoomed in by the submission system during the submission. 3. The content in the subtitles of each section of the paper needs to be enriched with corresponding descriptive text, rather than just one or two sentences. For example, "Circulation Count", "Player", "Bought Price" and so on in Section 2.2.1 need to be enriched. We have fixed the subheading for section 2.2.1. However, the mentioned phrases (“Circulation Count”, “Player”, etc.) are not subheadings but rather the names of the variables (explained after the semicolon) used in the model (and presented also in Table 2). ================================================================== More importantly, the content of the article needs to be modified, including the following points 1. The data used can be updated. This paper applies the transaction data from 07/27/2020 to 03/19/2021. You can consider applying the latest transaction data to this paper. We would like to thank the reviewer for this comment. We understand that using more recent data is something that can provide further value but there are a few reasons that we are not able to do so. The practical reason is related to the fact that the platform that provides the transactions (cryptoslam.io) is no longer open/free for use. While we could attempt to obtain the necessary funding to obtain a subscription to their API, the specific platform we are studying (i.e., TopShot), while still active, currently has nowhere near the volume of transactions that our dataset covers. In fact, the transactions declined drastically after 05/2021 (our data ends at 03/19/2021), and the year-to-year decline in the volume has been at 94%. The data we used in the paper includes the vast majority of the lifetime activity of the platform. Even though there is still trading activity on the platform, we believe that our dataset covers a large variety of the different users and transactions that have taken place on the platform. 2. The application of the conclusion is limited. According to the research conclusion of the paper, it is not possible to accurately evaluate whether a transaction is abnormal. We completely agree with the reviewer that we cannot evaluate the exact accuracy of the detection. However, this is the case with anomaly detection systems for the majority of the cases, where ground truth is not available. Now for some – well established – tasks/problems (e.g., spam detection) there are benchmark datasets with ground truth that can be used. However, in our case this is clearly not the case. Furthermore, the purpose of “fraud analytics tools” is to build models that will help analysts flag fraudulent activity to be investigated further. This further investigation is usually not automated but rather manual (similar to how IRS audits tax returns – either randomly or after triggers from the system). We have updated the conclusions in the revised manuscript to make this point even more explicit. 3. The paper does not consider the user's transaction logic and transaction behavior, but only considers the static transaction price and other information. You can also consider building the behavior model of transaction users/agents. This is certainly a very interesting idea, in that we could create a behavioral model for a user, including “rules” on how they make decision on buying and selling their NFTs. We could then simulate the trading process using this model and look at discrepancies between the actual trade data and the simulations. However, this would again have the limitation of identifying as anomalous the transactions that do not “fit” in the behavioral model we consider. We could possibly learn some of these rules by designing experiments to obtain observations of trades in such a way that will allow us to isolate various behavioral components. However, this is not possiblee and our expertise and strengths do not lay in the design of experiments and behavioral modeling. Furthermore, our pricing model can be seen as a behavioral model – to an extent - with respect to the decision on the price a seller is willing to sell, and a buyer is willing to buy a specific NFT. The various model parameters capture the variables that traders are considering when buying/selling. Nevertheless, this model is limited in only including parameters available in our dataset. 4. In the result analysis part, the probability threshold (1%) selected in the paper lacks the necessary theoretical support and detailed description. The probability threshold in our framework is very similar to the p-value in traditional hypothesis testing. Specifically, it represents the probability threshold below which we consider the chance of the price having that high/low of a residual as too small to not be anomalous. This probability controls essentially the type I and II errors. Having a higher threshold (e.g., 10%) we will flag more transactions as anomalous and potentially illicit, but a large fraction of them will be “false positives”. With a lower threshold (e.g., 1%) our filtering is more conservative, and we will flag less transactions as possibly illicit with a smaller percentage of them being false positives. However, it is possible that we will have missed flagging more true illicit transactions (i.e., increased “false negatives”). Similarly, with statistical hypothesis testing the choice of threshold is very subjective – and typically, and arbitrarily, selected as 10%, 5%, or 1%. While we have chosen the threshold to be 1%, the most appropriate way for the framework to be used is to output the actual probability of the transaction not fitting a “normal” profile according to the model and let the end user/auditor decide whether this is “small enough” to warrant a flag and further audit. 5. It can be considered further about the underlying blockchain technology corresponding to NFT, such as smart contracts, to enrich the judgment method and conclusion interpretation of abnormal transactions. We would like to thank the reviewer for this comment, which indeed can help in detecting abnormal transactions. However, the issue with adopting some of these features in our framework is the fact that our dataset does not include any information related to the smart contract protocols. More importantly, the platform itself – even though it uses a blockchain – is a centralized marketplace, and it is unclear on whether there is a practical use of the various benefits that the underlying technologies can provide. However, in a different system that is really based on a distributed ledger with smart contracts, features obtained by the various protocol variables can be extremely helpful, possibly in the auditing of the anomalous transactions identified by the “first level” of anomaly detection. We have provided additional details on this in the revised manuscript. There are still quite a few problems in this paper, but the research topic of this paper is of significance. After further revision, I hope this paper can become a good academic paper. Reviewer #2 The paper is written in a clear and concise manner, with proper technical terms used where necessary and no grammatical errors. The objective and methodology are clearly stated and easy to understand. The topic of NFT markets and their potential for illegal activities is highly relevant and timely, and the focus on the NBA TopShot marketplace is unique and appropriate. The methodology used to build the framework for labeling transactions as anomalous or not is well-explained and appears to be appropriate for the problem at hand. 1. The results of the network analysis are well presented and provide evidence for the authors' conclusions. However, it would be helpful to have a further explanation of the network metrics used and how they support the findings. We would like to thank the reviewer for this comment. We have added some details and information about the various metrics that we used that we believe will help the readers that are not familiar with network science understand them better. 2. The abstract mentions that the transactions labeled as anomalous will need to be further audited, so it would be helpful to discuss potential avenues for future work. We would like to thank the reviewer for this comment, and accordingly we have updated the conclusions in the revised manuscript to provide some more on what this further auditing might entail, which can also serve as a roadmap for future work. Attachments Attachment Submitted filename: plos-nft-rebutal.pdf https://doi.org/10.1371/journal.pone.0287262.r002
2 Jun 2023 Decision Letter - Junhuan Zhang, Editor Spotting Anomalous Trades in NFT Markets: The Case of NBA Topshot PONE-D-22-31512R1 Dear Dr. Pelechrinis, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Junhuan Zhang, PhD Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: (No Response) Reviewer #2: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ****** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: (No Response) Reviewer #2: (No Response) ****** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: Yes: Wenbin Wu ******** https://doi.org/10.1371/journal.pone.0287262.r003
Formally Accepted
6 Jun 2023 Acceptance Letter - Junhuan Zhang, Editor PONE-D-22-31512R1 Spotting Anomalous Trades in NFT Markets: The Case of NBA Topshot Dear Dr. Pelechrinis: I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. If we can help with anything else, please email us at plosone@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Junhuan Zhang Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0287262.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .