Wikipedia:Reference desk/Archives/Computing/2014 August 9

Computing desk
< August 8 << Jul | August | Sep >> August 10 >
Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


August 9

edit

Windows installer

edit

I am running vista on a laptop and when i try to update my antivirus program, it says it cant find the Windows installer. I have tried downloading all the Windows installers from Microsoft and each one says that it does not apply to my system. What is the correct installer for Vista and how can I get it working. Thanks for any suggestions/help--86.180.139.25 (talk) 12:18, 9 August 2014 (UTC)[reply]

Tweaking.com has a repair tool that has worked for me in the past.[1] --  Gadget850 talk 13:23, 9 August 2014 (UTC)[reply]
Just tried that and after about an hour of it fixing everything, yes i can once again update my antivirus. Thank you.--86.185.245.35 (talk) 20:49, 9 August 2014 (UTC)[reply]

Does Knowledge representation and reasoning has any real-life application?

edit

Like some Photoshop of KR that solves problem x? OsmanRF34 (talk) 16:27, 9 August 2014 (UTC)[reply]

Absolutely. I believe vibration analysis was one of the earliest uses. An expert can tell you that "this vibration frequency and magnitude, at this location on the machine, means you need to replace this part". A computer can be programmed to do the same analysis. StuRat (talk) 18:34, 9 August 2014 (UTC)[reply]
Yes, but would you need to design explicitly what state implies 'this machine part is damned'? Can't you just let a computer definy by itself what concrete situation is screwed through machine learning and some sort of feedback? Can a human perform a better job here than a computer analyzing and defining a problem with possibly an unknown number of variables?OsmanRF34 (talk) 21:24, 9 August 2014 (UTC)[reply]
I would expect that machine learning would eventually work, but you'd have a lot of damaged equipment before it got it right. Better to use the experience of the expert as a starting point, and maybe tweak the analysis from there: "The expert said that vibration magnitudes over X db mean we need to replace part Z, but wear analysis of the replaced parts has since shown that we can wait until 1.3X db to replace that part." StuRat (talk) 00:53, 11 August 2014 (UTC)[reply]
Do you know about Ontology_languages and the semantic web? Have you ever used a good search utility at a library? Have you ever had Netflix or Amazon or Facebook suggest something you might be interested in? These things are all related. Certain aspects of the nomenclature are still in flux, but KR is basically an AI spin on things like information organization and information retrieval. I suspect you've used many tools that are informed by study of these areas... SemanticMantis (talk) 21:06, 11 August 2014 (UTC)[reply]
I am not sure we need a human to define an ontology or at least mark what he thinks its important and then and only then tell the computer to compute what is important. we can let the computer discover what is related to what. Indeed, I think that letting a human define an ontology could be detrimental since the solution could be completely non-intuitive. OsmanRF34 (talk) 21:17, 11 August 2014 (UTC)[reply]
Well, if you want to let the computer do it, then you're back to ideas more similar to clustering algorithms and machine learning, etc. I just wanted to provide you with some links and context you might not have had. The field of information organization is growing quickly, drawing from computer science, logic, applied math, and library science. There has been a lot of hype about some of these ontologies and semantic webs. If you want to look for applications today, that's where it's at. And indeed, a human-built ontology language turns out to be a very useful thing for managing a large catalog of products, a museum, a body of scientific research, etc. Just don't fall into the trap of thinking that you know more about what tools are appropriate and useful than the researchers who have been studying these problems for decades ;) SemanticMantis (talk) 15:37, 12 August 2014 (UTC)[reply]