Comments on: What Comes After Google? http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/ Fri, 04 Jul 2008 17:15:17 +0000 http://wordpress.org/?v=MU By: Search and Social Networking « http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-2077 Search and Social Networking « Mon, 05 Nov 2007 17:48:03 +0000 http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-2077 [...] of a search tool is to help you find the information you need with the least effort possible (see Relevance/Effort metric). To this end, if there is someone who has, and is willing to share, the information or expertise [...] [...] of a search tool is to help you find the information you need with the least effort possible (see Relevance/Effort metric). To this end, if there is someone who has, and is willing to share, the information or expertise [...]

]]> By: Toufique http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-1420 Toufique Wed, 10 Oct 2007 19:47:21 +0000 http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-1420 Hi Pat, You're absolutely right. NLP is a fascinating area of AI research that holds a lot of promise if applied correctly. I'll dissect it using the same framework: <i>Avg. Relevance:</i> Using steps 3-5, NLP can certainly improve the avg relevance by understanding the actual "meaning" of a page, instead of just checking for the presence of a keyword. <i>Recall Effort:</i> I don't see NLP vastly improving keyword query entry because keyword search has already conditioned us to enter as few keywords as possible to describe what we're looking for. Users don't want to type in "When is Able Lincoln's birthday?" when "Abe Lincoln birthday" works just fine. Whenever we see a search box, grammar tends to go out the window. I think the true potential of NLP as a query mechanism will be realized after speech recognition matures. NLP has a huge opportunity to reduce recall effort for spoken queries. That should be very interesting. Toufique Hi Pat,

You’re absolutely right. NLP is a fascinating area of AI research that holds a lot of promise if applied correctly. I’ll dissect it using the same framework:

Avg. Relevance: Using steps 3-5, NLP can certainly improve the avg relevance by understanding the actual “meaning” of a page, instead of just checking for the presence of a keyword.

Recall Effort: I don’t see NLP vastly improving keyword query entry because keyword search has already conditioned us to enter as few keywords as possible to describe what we’re looking for. Users don’t want to type in “When is Able Lincoln’s birthday?” when “Abe Lincoln birthday” works just fine. Whenever we see a search box, grammar tends to go out the window. I think the true potential of NLP as a query mechanism will be realized after speech recognition matures. NLP has a huge opportunity to reduce recall effort for spoken queries. That should be very interesting.

Toufique

]]> By: Pat http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-1401 Pat Wed, 10 Oct 2007 02:02:50 +0000 http://youlicit.wordpress.com/2007/10/04/whatcomesaftergoogle/#comment-1401 What about an intelligent agent that can: 1. understand the query using NLP 2. ask clarifying questions and/or present sub-categories to follow 3. use nlp on the web pages to understand the meaning 4. summarize from numerous web sources to answer the question 5. point the user to the source docs the summaries came from Admittedly we are a ways away from anyone demonstrating this but many are working on it. Pat What about an intelligent agent that can:
1. understand the query using NLP
2. ask clarifying questions and/or present sub-categories to follow
3. use nlp on the web pages to understand the meaning
4. summarize from numerous web sources to answer the question
5. point the user to the source docs the summaries came from

Admittedly we are a ways away from anyone demonstrating this but many are working on it.

Pat

]]>