Comments on: SIGIR 2010: Day 3 Industry Track Keynotes

By: Daniel Tunkelang

Daniel Tunkelang — Wed, 04 Aug 2010 15:05:46 +0000

The Yandex presentation emphasized the need for geo-specific ranking because relevant result sets vary across all regions and countries. Consider a query for "pizza". Returning locally relevant results results is a lot easier if you have some idea of where the user is and what are the nearby pizza places.

By: Itman

Itman — Wed, 04 Aug 2010 14:52:37 +0000

What I’m surprised he didn’t mention is the challenges of data acquisition–in general, for domains beyond the web, obtaining high-quality data is often a much bigger challenge than filtering and ranking it.

Could you elaborate on this? Yandex is a web search engine. What kind of challenges should it mention then?

By: SIGIR 2010: Day 3 Industry Track Afternoon Sessions

SIGIR 2010: Day 3 Industry Track Afternoon Sessions — Wed, 28 Jul 2010 01:46:16 +0000

[...] DISCLOSURE ← SIGIR 2010: Day 3 Industry Track Keynotes [...]

By: Блог компании Яндекс / [RSS-пост] Яндекс на SIGIR 2010 « Мой новый блог

Блог компании Яндекс / [RSS-пост] Яндекс на SIGIR 2010 « Мой новый блог — Tue, 27 Jul 2010 05:50:11 +0000

[...] несколько человек написали про выступление (например, здесь и [...]

By: Daniel Tunkelang

Daniel Tunkelang — Mon, 26 Jul 2010 18:42:54 +0000

You can check out Yossi's presentation here. It's not just about integrating context, but in many cases organizing results around contexts.

By: jeremy

jeremy — Mon, 26 Jul 2010 18:35:36 +0000

Yossi’s presentation on “Search Flavours at Google” was a tour de force of Google’s recent innovations in the search and data mining space. The search examples most focused on the challenges of incorporating context into query understanding–where context might involve geography, time, social network, etc. But some of the more impressive examples showed off using the power of data to predict the present. More than anything, his presentation made clear that Google is doing a lot more than returning the traditional ten blue links.

Sorry, I missed Yossi’s presentation, so could you clarify: How exactly does integrating social and geographical context, as well as predicting the present, mean that Google has gone beyond 10 blue links?

What I mean is, you can integrate prediction and context, and still only show the user 10 blue links. Those 10 links will be higher “quality”, because you’ve integrated that additional information. But using social/geographic/etc context alone doesn’t automatically mean that you’ve gone beyond 10 links.

Or what am I not understanding?