bitcake

Disclaimer: As this account of plagiarism has not been confirmed by a legal authority, please interpret this article as an allegation rather than established judgment.

Introduction

Usually when you hear about plagiarism in school, a student has tried to pass off somebody’s work as her own. The school notices the similar work and disciplines the student accordingly. Punishment can range from a failing grade on the assignment to more serious repercussions like failing the class or even expulsion.

This, however, is not your commonplace case of a student using someone else’s work. This is a tale of a teacher and student taking the shared work of another student. And unlike the typical stern response to student plagiarism, the school doesn’t seem to care when one of its own teachers is doing the stealing.

Story at a Glance

1. I wrote a research paper with a classmate for a University of Central Florida (UCF) graduate course.
2. Same classmate later publishes a remarkably similar paper with the professor of that class without my consent or acknowledging my contributions to the work.
3. When I contact UCF and the publisher about plagiarism, both deny the allegations.

Reference

Scene Tagging - Original Paper (annotated) (2185 downloads ) Scene Tagging - Published Paper (annotated) (1692 downloads ) Scene Tagging - Published Paper w/ acknowledgments (2010 downloads ) Scene Tagging - Published Paper w/ updated author (1689 downloads )

Peter Matthews, [Andrew Mantel,] Cliff C. Zou. “Scene Tagging: Image-Based CAPTCHA Using Image Composition and Object Relationships”, in Proceedings of ACM Symposium on Information, Computer and Communications Security (ASIACCS), p.345-350, Beijing, China, 13-16 Apr. 2010.

The included numbered annotations are used to highlight the similarities between the two papers. These were sent to the involved parties. You may need to enable annotations in view options to see them.

My Story

The original work in question was a term project for the spring 2009 University of Central Florida (UCF) graduate course CAP6135 Malware and Software Vulnerability Analysis for which Cliff Zou, Associative Professor of the Department of Electrical Engineering and Computer Science, was the teacher. Myself (Andrew Mantel) and fellow student Peter Matthews conceived, developed, and wrote a research paper based on the concept of “scene tagging” for this project. It was about a year later that I discovered a strikingly similar paper had been published with Association for Computing Machinery (ACM). This would normally be cause for celebration, except that while my classmate’s and professor’s names were on the paper, my own name was astonishingly absent.

The paper published by Matthews and Zou lacks any mention of my name or my contributions to this unique concept, yet it takes direct elements from the original work such as the basic concept of scene tagging, spatial relation based scene tagging, and object quantity based scene tagging.

I contacted Bernard Rous, Director of Publications at ACM, with an allegation of plagiarism. I included both the original and published works in my communications, highlighting obvious similarities and paraphrasing between the two (see above for annotations). Mr. Rous, however, denied that this was plagiarism for the following reasons:

1. “one of the authors was himself a co-author of the earlier work. As a co-author, he can re-use his earlier material.”
2. “the earlier work was unpublished and does not require citation”
3. “an acknowledgement has been made to your work… ACM received the request to insert an acknowledgement in the paper prior to receiving your allegation of plagiarism.”

My responses to these points:

1. “If he was the sole author of the earlier work then it would be reasonable that he could re-use his earlier material; however the earlier work was a group effort. Both of our names are at the top of the earlier work, once submitted one cannot discriminate who provided what content. We both owned the earlier work, so I don’t think only one of us can re-use it without citing the other contributor.”
2. “I wasn’t sure about this either until I read ACM’s plagiarism policy on http://www.acm.org/publications/policies/plagiarism_policy where they state:
   ‘Additional information regarding how to obtain involved papers that are unpublished (e.g., a technical report, an online posting).’
   From my view this means unpublished papers are suitable for plagiarism inquiries, otherwise why mention it at all in your policy?”
3. “I sent UCF a plagiarism inquiry on 12/12/2010 and ACM an inquiry on 12/17/2010. From my correspondence with UCF I didn’t think Professor Zhou contacted ACM with an update for my acknowledgement until well after my inquiry with UCF. From what UCF forwarded me, Professor Zhou made the request around 1/28/2011 and only because UCF contacted him about it, meaning it was only after I contacted UCF.”

Despite my protests, ACM would not reconsider their decision.

I also contacted UCF’s Office of Student Rights and Responsibilities. I believed that UCF, which heavily instills upon its students the importance of academic honesty, would treat my case with the utmost responsibility. After a month of communications between myself, UCF, and Professor Zou, the published paper was updated with the following acknowledgment section:

This research work was initially started from a course term project “Image and Scene Tagging: An Alternative Approach to Current CAPTCHA Techniques” conducted by graduate students Andrew Mantel and Peter Matthews in Spring 2009 in University of Central Florida.

An acknowledgment section at the end of the paper? That is not the proper citation when paraphrasing sections of another’s work. It also completely ignores the initial act of plagiarism. I continued corresponding with UCF, asking them to properly recognize this case as plagiarism and treat is accordingly, but to no avail. After several more months of emails, my final communication with UCF was a phone conversation with Dr. Charles Reilly, Professor and Associate Dean for Academic Affairs, and Patricia MacKown, Director of the Office of Student Rights and Responsibilities. I took the following notes during the conversation:

• Dr. Reilly does not consider this plagiarism.
• I asked if even when a sentence is repeated pretty much verbatim, shouldn’t that count as plagiarism, but Dr. Reilly said it does not. I used highlight 12 as the main example throughout the discussion. I repeated this question multiple times because I could not believe what I was hearing, yet every time the response was that copying a sentence does not count as plagiarism.
• I asked if the situation were reversed, had the published paper appeared first and I had written the original paper and submitted it to UCF without citation, would I be in trouble for plagiarism. Dr. Reilly said that it would not necessarily be plagiarism.
• Dr. Reilly noted that of the highlights I submitted, only highlights 10, 11, and 12 were of value. I partially agreed with him but stated that the other highlights were simple demonstrations that the published paper as a whole is a paraphrase of the original work.
• Dr. Reilly says it is common to continue research on a concept even if the original author is not involved. I responded that this was not a continuation of research, it was a rephrasing of what had been written before. In other words, not much new was added to separate the two works. Also, when you continue a work, you still have to acknowledge the earlier contributors.
• UCF thinks that adding the acknowledgments section a year later was a good enough response. I told them that adding an acknowledgments section is not the proper citation when paraphrasing from another work, but Dr. Reilly said this was common and acceptable.
• They claimed that this case has been passed around to a lot of faculty at UCF. Dr. Reilly says that plagiarism is up to interpretation, that some people would disagree with him, but that he believes this is not plagiarism. I do not know who else, if anyone, saw this case. I stressed that this is not a matter of interpretation, this is blatant plagiarism.
• Dr. Reilly used an example that later that day he had a meeting about a student who submitted a paper where five to six pages were copied directly from a website. I told him that I understand that is a more obvious case of plagiarism, but that doesn’t change the fact that this is plagiarism too even if entire pages were not copied verbatim.
• Dr. Reilly said it is common for professors to simply attach their name to a paper being published. I responded that even if that is common practice, it is no excuse. If your name is at the top of a paper, then you are responsible for it.
• I told them that there was no excuse for Professor Zou not to realize that this material was plagiarized from the original paper since that first paper was created for the course we was the professor of. Zou had closely read the original paper before since he had to grade it, and as such had intimate knowledge of it. Dr. Reilly replied that it had been a year between my submitting the original paper to UCF and the paper being published so Zou likely forgot about the original work. I again stressed that there is no excuse for forgetting the original work since he obviously had knowledge of it.
• Ms. MacKown read the definition of plagiarism as per UCF’s Golden Rule:
  “Plagiarism: Whereby another’s work is used or appropriated without any indication of the source, thereby attempting to convey the impression that such work is the student’s own.”
  After hearing that I said that this is exactly what is occurring here, but Dr. Reilly disagreed.
• Dr. Reilly said they would use this case as a wake up call for professors. But if this is a wake up call, wouldn’t that mean it’s an example of wrong doing?
• Dr. Reilly stated he will not contact ACM since he does not think this is plagiarism.
• UCF claims they have never received notice from ACM about this case.
• I told them we’ve reached the end of what can be achieved with conversation between myself and UCF, but I will continue to pursue this matter.

In the end, UCF would not agree that this is plagiarism. I eventually received word from Professor Zou that the published paper had been updated again, this time with the acknowledgment section removed and my name appearing at the top of the paper as a co-author. Seriously, in what case of “not plagiarism” is a person’s name added as an author?

When I requested documented reasons from Dr. Reilly on why this case should not be treated as plagiarism, he only provided me with one:

If Mr. Matthews were not a co-author of both of these works, I may have reached a different conclusion. Mr. Matthews is no less entitled to use the work you did together on your course project than you are.

ACM used a similar excuse. Somehow these institutions think it is acceptable for an individual to take sole credit for the work of the group.

What Is Plagiarism?

The definition of plagiarism seems obvious to anyone but can be hard to put into words. I like to think of it this way: If Work A could not exist without Work B, then Work A should properly acknowledge Work B’s influence. In early grade school this usually means putting a person’s name after a direct quotation, but this concept also applies to paraphrasing and expression of ideas.

From what I’ve read about copyright, it protects not only published works but anything written. Even something jotted down on a bar napkin is protected. Now, if that napkin is never seen by anyone and someone else comes along with a similar writing, that is not a copyright issue. In my case though, the paper submitted to UCF as a school project was well known to both the classmate that co-wrote it and the professor that graded it, along with the rest of the class we presented it to.

Another key point to note is that copyright protects expression of idea rather than the idea itself. Although it may not seem honorable, the fact that the published paper is about the same idea of “scene tagging” may not technically be the problem here. Plagiarism occurs when ideas about scene tagging in the published paper are conveyed in a similar manner to the original paper, as demonstrated below.

For more information on US copyright, visit http://www.copyright.gov/help/faq/faq-protect.html

Examples

The following are some of the highlights I used when communicating with UCF and ACM to showcase the similarities between the two papers. For a full list of examples, check out the annotated papers.

Highlight 10:

Original paper:

Distortion of the composite image, conversely, primarily attempts to make machine recognizability of objects more difficult. However, preserving the recognizability of the objects placed in the scene is of the utmost importance for the system to be able to create scene instances which humans may answer. Thus, a careful choice of distortion effects and parameter ranges must be chosen as to maximize the difficulty it causes for machine object recognition while not distort the objects in the image beyond the point of human recognition.

Published paper:

Distortion of the composite image, conversely, primarily attempts to make machine recognition of objects more difficult. At the same time, human recognition of these objects is crucial. Thus, the sequence of distortions must be chosen as to make things difficult for automated attacks without distorting the objects in the image beyond the point of human recognition.

Highlight 11:

Original paper:

they add elements of interest to portions of the background image that may otherwise be uninteresting… This ensures that a flat portion of the background does not make the task of isolation of an object in that portion of the image significantly easier than it would if placed elsewhere.

Published paper:

it may add elements of interest to areas of the background image that may otherwise be uninteresting. This not only makes object isolation more difficult, it also results in a larger number of extraneous image interest points.

Highlight 12:

Original paper:

The use of random shape placement in this stage attempts to exploit the Gestalt perception abilities of humans, namely that human have a strong ability to recognize and understand images in the face of incomplete or fragmented visual information whilst machines have a very difficult time doing the same thing.

Published paper:

The use of clutter in this stage attempts to exploit the Gestalt perception abilities of humans, namely that human have a strong ability to recognize and understand images in the face of incomplete or fragmented visual information while machines have a difficult time doing the same thing.

Overlap:
“The use of [random shape placement/clutter] in this stage attempts to exploit the Gestalt perception abilities of humans, namely that human have a strong ability to recognize and understand images in the face of incomplete or fragmented visual information [whilst/while] machines have a [very] difficult time doing the same thing.”

These are almost identical expressions of the same ideas. According to UCF and ACM, this is not plagiarism >_>

What Do I Want?

When I began my plagiarism inquiries, I had two major goals in mind.

First, I wanted the published paper to properly acknowledge my contributions to the work. After many months of struggle, this goal was accomplished when the published paper was updated with my name as a co-author.

Secondly, I want the act of plagiarism to be properly recognized. If a person robs a bank, the thief caught months later cannot simply give the money back without facing condemnation and punishment. I want the plagiarizers involved to receive appropriate admonishment. Their names should be flagged in the professional community as having committed plagiarism. I would have liked the paper flagged on ACM as being plagiarized, but at this point that may be tricky since it now includes my name as a co-author. I assume there are special chastisements that occur for members of a public university.

Admission of plagiarism has yet to be realized and, based on my experience, probably never will be.

Thank You

Thank you for taking the time to read this article. I have been deeply hurt by this incident. Not just because my former teacher and classmate plagiarized me, but that no authority would admit to it when brought to their direct attention.

Please feel free to share this article. And be careful not just to avoid plagiarism, but to make sure you yourself are not plagiarized.

Written by: Andrew Mantel. April 16, 2013.