home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.ai      Awaiting the gospel from Sarah Connor      1,954 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 1,399 of 1,954   
   Greg Tchil to gurpreet singh   
   Re: AI in a website   
   01 May 07 13:45:42   
   
   From: gtchill@gmail.com   
      
   "gurpreet singh"  wrote in message   
   news:457574da$1@news.unimelb.edu.au...   
   > Ted Dunning wrote:   
   >> gurpreet singh wrote:   
   >> > is there no one who can reply me .........................   
   >> >   
   >>   
   >> Your question was so non-specific as to be unanswerable.   
   >>   
   >> There are lots of examples of web-sites that use AI.  Google.  Pandora.   
   >>  Veoh.  Yahoo.   
   >>   
   >> So what?   
   >>   
   >> That just doesn't seem very intersesting.   
   >>   
   >   
   >   
   > ok this might uninteresting for you give me a chance i will make it   
   > look interesting to you   
   >   
   > but can you tell me how to accomplish that and where is the artificial   
   > intelligence in these sites . i know some about it like we have   
   > "spelling suggestions" in google search it is one example of AI in a   
   > website . can you give me some more examples about it and if you have   
   > any suggestions of yours it will  be great and one thing more please   
   > tell me how to accomplist the same i just dont want to know the algo   
   > please tell me the practical implementation of doing that if it is   
   > possible   
      
   Hi Gurpreet,   
      
   Actually, spelling corrections and other search suggestions can be done   
   without the use of AI methodologies. An advanced Algorithm or Statistics   
   class (MS level or above) might help you to unlock the "magic" that may seem   
   present in some of the things you see online these days.   
      
   My suggestion would be to produce a site that tries to predict what the next   
   page you will load will be. This should be based on the USERS OWN previous   
   selections, not by using simple statistical methods or data-mining on a   
   larger population. You would need a pretty substantial site to do this   
   (enough content and pages to keep it interesting). You could use the   
   predictions to queue content for faster delivery etc. It would be of little   
   practical use (in todays broadband world) but may have the "wow" factor you   
   are looking for.   
      
   Greg T   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca