In June of this year, Enamul Haque, a PhD student at the University of Waterloo, in Canada, came across an article in the International Journal of Advanced Computer Science and Applications (IJACSA).
It looked familiar.
That’s because it was copied, in large part, from Haque’s master’s thesis, which he had completed at Canada’s McMaster University and submitted the previous year. Haque wrote to Kohei Arai, the journal’s editor in chief, on June 30, providing detailed evidence of plagiarism:
I am Enamul Haque, completed my Masters (by thesis) last year from McMaster University under the supervision of Prof. Fei Chiang and currently pursuing my PhD at “Data Systems Lab” of University of Waterloo under the supervision of Prof. Grant Weddell and Prof. David Toman.
Few days ago, I came across a data science article titled “A Data Science Framework for Data Quality Assessment and Inconsistency Detection” (link) published in your Journal [(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 12, No. 4, 2021] (paper 76), and I am extremely upset to discover that there are almost all the sections, algorithms and even plots are unoriginal, and deliberately copied from the previously published work of mine: “Restoring Consistency in Ontological Multidimensional Data Models via Weighted Repairs” (link). I have consulted it with my thesis supervisor Prof. Fei Chiang at McMaster University, and she advised to notify the journal for further investigation and necessary actions.
There are some tools that can check plagiarism, we have tried one of those too (turnitin). As the paper was already been published, so we inserted our thesis into that system, and they produced the output (please follow this link) where the similarity with the paper has been marked as number one “1” which is the paper published in your journal.
In this case, the overlap goes beyond the normal occurrence of standard phrases in the field. I should say it is almost 100% copied from the thesis. Moreover, the thesis has not been cited and they did not even mention reproduced or asked for any permission from us. For this reason, we would like to request you formally to take necessary steps against this submission please.
In contrast to plagiarism cases that publishers often take years to adjudicate, the reaction was swift. The journal acknowledged Haque’s email the same day, and two days later, they wrote:
Thank you for bringing this to our attention. Indeed, this seems to be a case of Level 1 plagiarism – Uncredited verbatim copying of a full paper.
The published paper will be retracted with a publication of a notice of violation of publication principles, in the electronic database as part of the article’s bibliographic record at IJACSA.
As a corrective step, we shall issue a warning letter to the author with publication ethics and publication malpractice statements and will also be informing the head of the author’s institution.
Once again we appreciate you contacting and reporting this issue to us.
That email apparently never reached Haque, and when he wrote to check in on the case later in July, the editor re-sent it and added:
As per the email, all corrective actions were taken and the published paper was replaced with a retraction notice. In addition to that, all authors found to be a part of this malpractice were barred from all SAI publications for a period of 2 years.
Here’s the notice:
After careful and considered review of the content of this paper by a duly constituted expert committee, this paper has been found to be in violation of IJACSA`s Publication Principles. We hereby retract the content of this paper. Reasonable effort should be made to remove all past references to this paper.
The first author of the paper, Anusuya Ramasamy of Arbaminch University in Ethiopia, did not respond to requests for comment from Retraction Watch.
Like Retraction Watch? You can make a one-time tax-deductible contribution or a monthly tax-deductible donation to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
Nice to see action taken so quickly.
Now Haque should write up his own paper based on the thesis and submit it to the journal.
Or, he can just take the IJACSA paper and resubmit it under his own name! 😉
But not really nice to see what a general, non-specific retraction notice was published. Other than the statement that the paper is “in violation of IJACSA`s Publication Principles” (whatever those may be) it doesn’t specify the actual reason for the retraction (plagiarism). How nice for those authors that stole another person’s work that the publisher is willing to cover-up for the crime. No impressed at all.
“case of Level 1 plagiarism” — there’s now levels of plagiarism? I guess it doesn’t matter, as “IJACSA” itself doesn’t inspire any confidence. It looks like a typical predatory journal.
Don’t link to PDFs. Link to the page the PDF is on.
I don’t really get that. Beyond the website being a bit dated, what makes this journal ‘look like a typical predatory journal’? They just retracted an article more swiftly than most traditional publishers, they are indexed in various databases, and archive in PORTICO.
(IJACSA) International Journal of Advanced Computer Science and Applications, Vol. 10, No. 4, 2019
“A Comprehensive Collaborating Filtering Approach
using Extended Matrix Factorization and
Autoencoder in Recommender System”
“Deep semantic similarity model (DSSM) represents a broader model of semantic similarity [26]. DSSM, created by the MSR Deep
Learning Technology Center, is a profound neural system
(DNN) displaying strategy for speaking to content strings
(sentences, inquiries, predicates, substance specifies, and so
on.) in a consistent semantic space and demonstrating semantic
similitude between two content strings [27]. DSSM has wide
applications including data recovery and web look positioning,
promotion choice/significance, logical substance hunt, and
intriguing quality undertakings, question noting, information
derivation, picture, and machine interpretation. DSSM can be
utilized to create idle semantic models that venture elements
of various kinds (e.g., inquiries and archives) into a typical
low-dimensional semantic space for an assortment of machine
learning assignments, for example, positioning and order. For
instance, in web seek positioning, the significance of a report
given a question can be promptly figured as the separation
between them in that space.”
Do you feel the familiar language with “profound neural systems” and the like?
A retraction doesn’t mean anything – retracting one piece of trash they still have lots of them to offer.
More clearly fraudulent in this case even: ” […] a profound neural system (DNN) displaying […]”. The acronym DNN obviously means Deep Neural Network here.
They used the “tortured language” method to replace the original, correct term with “profound neural system” (which is meaningless gibberish), but forgot to change the acronym.
That should probably be recognised as a clear red flag in general, when a known, standard acronym is used with terms that don’t match it.
An interesting comparison is to the description of DSSM on the Microsoft Research website which reads in part:
DSSM stands for Deep Structured Semantic Model, or more general, Deep Semantic Similarity Model. DSSM, developed by the MSR Deep Learning Technology Center (DLTC), is a deep neural network (DNN) modeling technique for representing text strings (sentences, queries, predicates, entity mentions, etc.) in a continuous semantic space and modeling semantic similarity between two text strings (e.g., Sent2Vec). DSSM has wide applications including information retrieval and web search ranking (Huang et al. 2013; Shen et al. 2014a,2014b ; Palangi et al.2016), ad selection/relevance, contextual entity search and interestingness tasks (Gao et al. 2014a), question answering (Yih et al., 2014), knowledge inference (Yang et al., 2014), image captioning (Fang et al., 2014), and machine translation (Gao et al., 2014b) etc. DSSM can be used to develop latent semantic models that project entities of different types (e.g., queries and documents) into a common low-dimensional semantic space for a variety of machine learning tasks such as ranking and classification. For example, in web search ranking, the relevance of a document given a query can be readily computed as the distance between them in that space.
So the article pointed out is clearly using plagiarized text, processed though the familiar method of “tortured language” (Here is the original text: https://www.microsoft.com/en-us/research/project/dssm/).
I wonder what happened with the acronyms though. The acronym DLTC was removed, although it would still have been correct in the plagiarised text, but DNN was kept in even though it no longer fits the words.
The “DLTC” in the original was a hyperlink. Maybe the language torturing is done by software that removes links?
Another article copied text from multiple sources, the second author blamed the first author after pubpeer post (https://pubpeer.com/publications/54CDDA74073A372A406F625E3D875D) but the editor decided not to retract the paper.
Author response:
I have asked the journal to retract the paper – As the second author and PI, I was unaware of this and am extremely disappointed in the first author. I will check for plagiarism carefully and systematically in future work with any author.
I was informed that the Editor has read the paper and plagiarized sections and has decided the outcome: the instances of plagiarism will be corrected through a corrigendum, correctly quoting the sections from the relevant papers. I will link to the corrigendum once published.
I feel like he did nothing wrong but there were so things he could have done differently.