A shameless reblog from Retraction Watch:
“What do porn star Ron Jeremy, Max Weber and Michael Jackson have in common? Very little — except the three names appear in the list of references for a recent hoax paper by a group of Serbian academics who, fed up with the poor state of their country’s research output, scammed a Romanian magazine by publishing a completely fabricated article.
The paper is replete with transparent gimmicks — obvious, that is, had anyone at the publication been paying attention — including a reference to the scholarship of Jackson, Weber, Jeremy and citations to new studies by Bernoulli and Laplace, both dead more than 180 years (Weber died in 1920). They also throw in references to the “Journal of Modern Illogical Studies,” which to the best of our knowledge does not and never has existed (although perhaps it should), and to a researcher named, dubiously, “A.S. Hole.” And, we hasten to add, the noted Kazakh polymath B. Sagdiyev, otherwise known as Borat.
The paper, “Evaluation of transformative hermeneutic heuristics for processing random data,” by Dragan Djuric, Boris Delibasic and Stevica Radisic, appeared in the magazine Metalurgia International, according to the website In Serbia, which reported on the story. The authors, from the University of Belgrade and the Health Center ‘Stari Grad’, appear on the manuscript in false wigs and mustaches.
Here’s the abstract from the article, in all its glorious meaninglessness:
The improved understanding and proper application of simulation models for various domains, from e-government to e-learning is an appropriate riddle. In this significant paper, we increasingly understand how randomized heuristic algorithms could be unexpectedly applied to the intuitive processing of random data in a novel way. While such a claim might seem counterintuitive, it is supported by prior relevant work in this thriving field. We describe a robust conceptual tool for solving this promising challenge using transformative hermeneutic heuristics for processing random data. Accordingly, the main focus of our work is, obviously, the evaluation of such methodology on an encouraging and intriguing subject of finding in which ways people in an insufficiently developed country see the aid provided by European Community. This illustrative case clearly demonstrates our profound approach, and, thusly, is a compelling foundation for future improvements of the methodology. In fact, the main contribution of our work is that we argue that although a random process might carry a slight risk of being insufficiently relevant for the problem at hand, the solution to any such conundrum could be surely looked for in a multidisciplinary approach
If this sounds like the work of Alan Sokal, it should. The Serbians tip their wigs to Sokal, whose 1996 mock paper Social Text, “Transgressing the boundaries: towards a transformative hermeneutics of quantum gravity,” caused the journal substantial chagrin.
A litany of previous work supports our use of self-learning archetypes . Our heuristics is broadly related to work in the fieldof hermeneutics by Sokal [Error! Reference source not found.], but we view it from a new perspective: random theory. 
We highly recommend reading the article, whose endless inside jokes make it effectively an infinite jest. In it you can find gems like this modest contextualization:
Our work has been inspired and directly founded on various astonishing research by intellectual giants in various interesting fields of social science and practically conducted and supported by the advances in multiple technical disciplines, thus giving this work a veritable multidisciplinary aura. We place our work in context with the prior work in several multidisciplinary areas.
Or this circular figure caption:
The decision tree model proposed in this paper is shown in Figure 7. It clearly presents the proposed model, which might be useful to EU analysts, but also to theorists who might judge the validity of this model using the new proposed heuristics.
As we will soon see, the goals of this section are manifold. Our evaluation could represent a valuable research contribution in and of itself. The first experimental results came from 2500 trial runs, and were not reproducible. The next batch of results come from only 50 trial runs, and were not reproducible. Continuing with this rationale,the many discontinuities in the graphs point to improved precision introduced with our decision tree algorithms. Such a hypothesis at first glance seems unexpected but fell in line with our expectations. As hypothesized, the final run was sufficiently consistent, which shows the useful convergence of our heuristics. Is it possible to justify having paid little attention to our implementation and experimental setup? Yes, but only in theory. Our evaluation strives to make these points clear.
According to In Serbia, the Romanian magazine
which is otherwise full of Serbian authors, published it in its entirety, without a single correction.”
appropriate music of mirth: