# Hase Occam's Razor ever been scientifically verified? **URL:** \ **Category:** Factual Questions\ **Created:** [October 27, 2003, 7:31pm UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894 "2003-10-27T19:31:07Z")\ **Posts on this page:** 5\ **Page:** 3
**Author:** ![DreadCthulhu](https://avatars.discourse-cdn.com/v4/letter/d/db5fbb/32.png) [@DreadCthulhu](https://boards.straightdope.com/u/DreadCthulhu)\ **Post date:** [October 29, 2003, 6:07am UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894/41 "2003-10-29T06:07:22Z")
> [@](#): > > \*Originally posted by Alan Smithee \* > \*\*There are an infinite number of unobserved phenomena that could be posed as ad hoc explanations for crop circles (faries, ghosts, angels, Cthulu, black helecopters, etc.), \*\* Just for the record, I DO NOT create crop circles, so you can rule that out. ---
**Author:** ![Triskadecamus](https://avatars.discourse-cdn.com/v4/letter/t/b19c9b/32.png) [@Triskadecamus](https://boards.straightdope.com/u/Triskadecamus)\ **Post date:** [October 29, 2003, 6:09am UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894/42 "2003-10-29T06:09:14Z")
When I examine the elements of my own thinking, and avoid invoking elements I cannot explain, or demonstrate, I am using a philosophic tool, to my advantage. When I try to object to your thinking because it invokes elements I cannot explain or demonstrate I am committing an error. Clearly, adding you to the hypothesis needlessly multiplies the entities. The Barber of Occam shaves only himself. Tris ---
**Author:** ![Shalmanese](https://avatars.discourse-cdn.com/v4/letter/s/45deac/32.png) [@Shalmanese](https://boards.straightdope.com/u/Shalmanese)\ **Post date:** [October 29, 2003, 6:11am UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894/43 "2003-10-29T06:11:46Z")
Perhaps a lot of people are being misguided by my cite since Occam’s Razor in Data mining terms has a quite precise and subtly different usage from common vernacular. It is used as a rule of thumb to prune areas of your decision tree that are unlikely to yield meaningful results. However, This researcher proved that this heuristic, which was just assumed to be true by a large part of the academic community, was in fact, largely worthless. In this case, Occam’s razor was emperically proven to be worthless. Similarly, in real life, if you could find a BETTER heuristic than Occam’s Razor, then it would also be proved to be false. ---
**Author:** ![KidCharlemagne](https://avatars.discourse-cdn.com/v4/letter/k/ed8c4c/32.png) [@KidCharlemagne](https://boards.straightdope.com/u/KidCharlemagne)\ **Post date:** [October 29, 2003, 6:29am UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894/44 "2003-10-29T06:29:00Z")
> [@](#): > > \*Originally posted by Alan Smithee \* > \*\*I’m not sure I agree with this, **KidCharlemagne**. That sounds like exactly the sort of misapplication that the researcher in **Shalmanese** ’s link tried to use. The “common sense” application is pretty clearly what Ockham had in mind. What’s hard to realize is that it wasn’t always common sense. Nor is it today! \ \*\* I do agree that the “common sense” application is probably what Sir Occam originally intended but Occam’s Razor has taken on a life of it’s own and I was trying to give the OP a sampler of how OR is interpreted/used today - interpretations that I think might make the concept worthy of such a fancy term. The interpretation of OR that I illustrated with trading system development is commonly used today by designers of neural net, pattern recognition, and fuzzy logic algorithms. It is probably not a strictly correct interpretation of OR but it is a reasonable and useful extension of the idea. A search on Occam’s Razor and pattern recognition revealed the following site which may be of interest to you: [http://cgm.cs.mcgill.ca/~soss/cs644/projects/jacob/pattern-rec.html](http://cgm.cs.mcgill.ca/~soss/cs644/projects/jacob/pattern-rec.html) [http://cgm.cs.mcgill.ca/~soss/cs644/projects/jacob/interpretations.html](http://cgm.cs.mcgill.ca/~soss/cs644/projects/jacob/interpretations.html) I would say that the academic in **Shalmanese’s** link missed the point of OR entirely while the artificial intelligence guys are making an extension/corrollary of it. ---
**Author:** ![KidCharlemagne](https://avatars.discourse-cdn.com/v4/letter/k/ed8c4c/32.png) [@KidCharlemagne](https://boards.straightdope.com/u/KidCharlemagne)\ **Post date:** [October 29, 2003, 6:38am UTC](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894/45 "2003-10-29T06:38:21Z")
> [@](#): > > \*Originally posted by KidCharlemagne \* > I would say that the academic in **Shalmanese’s** link missed the point of OR entirely while the artificial intelligence guys are making an extension/corrollary of it. \*\* Sorry **Shalmanese** , I came back to finish my post after being afk and didn’t see you had made a qualifying post in the meantime. Just didn’t want you to think I was completely ignoring what you had to say. [Previous page](https://boards.straightdope.com/t/hase-occams-razor-ever-been-scientifically-verified/209894.md?page=2)