Recently I heard about a website called Spinbot.com that will "rewrite human readable text into additional, readable text" - it is an "automatic article spinner" tool. The site is designed for "bloggers, twitter users and online marketers", and is described as a "dream come true" to help rework content to help with SEO. In its self-promotion, Spinbot.com states that with a "single click you can turn your old blog post or website article into a completely new one, thereby doubling the payoff you get in return for the time and energy you have already invested into creating quality website content". Nothing wrong with doing this I hear you say - re-writing one of your own old posts or any old material is OK, it is your own content.
However, there is (from an academic point-of-view) a more sinister use of this clever tool to avoid plagiarism. Before I go further - I have not detected this as a problem with any of my own students. Could it work that if you copy a passage, say from a book or website, put it through a tool like Spinbot to change words - that it would beat a plagiarism detection tools like Turnitin?
To test this out I took one of my own short posts: "New @YouTube Video: How To... Perform a Wilcoxon Signed Rank Test (By Hand)" written just yesterday to see what it would look like after it was spinned by Spinbot. Before doing this I ran the post through Turnitin using default settings and you can see that it had a 16% similarity with two Internet sources. None of this was copied from the sources listed and you can see that they are simple terms being reused.
 |
| Click image to enlarge. |
 |
| Click image to enlarge. |
Well well well - a 0% similarity index even though I didn't write a single word! Spinbot has completely changed the text and has beaten Turnitin. I'm not surprised that this might be tempting to students. However, there are some tell-tale signs that a spinner has been used to mix words. The "Wilcoxon Signed Rank Test" (actual name of test) is transformed into the "Wilcoxon Marked Rank Test" (no such test). Also the original text "For some time I have noted a gap in my series of Statistics By Hand series" is transformed into "For quite a while I have noticed a crevice in my arrangement of Insights By Hand arrangement". Finally, "Ultimately it will tell the researcher if there is a significant difference between two data sets or not. Enjoy!" is transformed into "Eventually it will tell the scientist if there is a huge contrast between two informational indexes or not. Appreciate!"
While there is some clever stuff going on here, the second readings after Spinbot has been applied make for more difficult reading that is not natural. Anyone aware of tools such as Spinbot.com will immediately be suspicious, even those not so may question the second passage above. I'll certainly be more aware of this and will be watching out for it in future.
Comments are disabled for this post to block Spammers trying to sell me essay-writing services.
Please note: Opinions and comment expressed in this post (and all posts in this blog) are mine alone and do not in any way represent those of anyone else or any institution.
