My Photo

 

  • Subscribe with Kindle

« Win $1, 000, 021 by Helping Netflix | Main | Klostu »

October 05, 2006

Powerset: Update

Update: TechCrunch picks up the story. One thing that is getting a little fuzzy here is where NLP ought to be applied in search. Understanding the query is only one part - the other part (to me, the more interesting part) is in understanding the text in the documents. This also requires smarts for understanding the structure and layout of documents and the function of different document areas (navigational, content bearing, etc.). This point appears to be missed in the TechCrunch post (and in the comments by readers there).

Briefly, Barney has written up in detail his vision regarding Powerset, NLP in search and the restrictions of keyword based search. This is worth reading - and I'll post more on it later. I wanted to link this article to a recent post by Paul Kedrosky which expressed satisfaction with current search engines:

I have no idea anymore what "better" search would mean. I find pretty much everything I want now, and while natural-language processing always sounds great, improvements in how I submit searches do diddly for me.

My analogy is as follows: imagine interacting with a reference librarian, but you could only speak in 2-3 word statements. The reason this analogy is useful is that it points out that the search problem, when not constrained by the assumptions and expectations of the text box, is far richer and complex than we've been blinkered to believe.

Oh yes - I also wanted to quote Barney on a most excellent turn of phrase describing keyword based queries language, which he describes as: a grunting pidgin language. Excellent!

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c994053ef00d8342e3dc553ef

Listed below are links to weblogs that reference Powerset: Update:

» The Powerset Blogstorm: 1 week later from Barney Pell's Weblog
I wrote a week ago about how Powerset had become the subject of a blog storm, and shared my vision of natural language search. Little did I realize that the storm had barely started. One week later, there are now... [Read More]

Comments

The application of NLP to IR is not new and I'm wondering if there is prior work on NLP actually improving results; I cannot think of any. It's a fine story but truly natural language interaction with a system comes close to asking for natural language understanding which is a tall order. The recent success in the subdiscplines of information extraction and interaction are probably better avenues to explore improving and changing search.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Twitter Updates

    follow me on Twitter

    July 2009

    Sun Mon Tue Wed Thu Fri Sat
          1 2 3 4
    5 6 7 8 9 10 11
    12 13 14 15 16 17 18
    19 20 21 22 23 24 25
    26 27 28 29 30 31  

    Categories

    Blog powered by TypePad