Comments on: Assisted search http://www.scienceforseo.com/information-retrieval/assisted-search/ a bridge between worlds Sat, 05 Feb 2011 14:02:08 +0000 hourly 1 By: CJ http://www.scienceforseo.com/information-retrieval/assisted-search/comment-page-1/#comment-1213 CJ Sun, 03 Jan 2010 00:38:28 +0000 http://www.scienceforseo.com/?p=1173#comment-1213 Thank you for the comment and your insight :)<!-- Touched by JuLiA --> Thank you for the comment and your insight :)

]]> By: FindItByMe http://www.scienceforseo.com/information-retrieval/assisted-search/comment-page-1/#comment-1211 FindItByMe Tue, 29 Dec 2009 11:39:29 +0000 http://www.scienceforseo.com/?p=1173#comment-1211 I wish I knew about this website and your research before. I'm an amateur with no formal training in linguistics and information retrieval, but as an enthusiast I noticed that people quickly adopt strategies for using search engines and these become reflexive and unconscious. Informally I've proved this to people who say that Google is easy to use; when I ask them to slow down, they realize how much they reformulate, backtrack and sometimes start over with a whole new search. But this exact scenario is also played out when people communicate. There are innumerable "clarification exchanges" that take place when one person is trying to understand what another wants. Your conclusion that the future of search needs to be in glowing neon lights. The conceptual jump from searches "rendering a list of links" to that final summary paragraph of requested information should be every search engine's goal.<!-- Touched by JuLiA --> I wish I knew about this website and your research before. I’m an amateur with no formal training in linguistics and information retrieval, but as an enthusiast I noticed that people quickly adopt strategies for using search engines and these become reflexive and unconscious. Informally I’ve proved this to people who say that Google is easy to use; when I ask them to slow down, they realize how much they reformulate, backtrack and sometimes start over with a whole new search. But this exact scenario is also played out when people communicate. There are innumerable “clarification exchanges” that take place when one person is trying to understand what another wants.

Your conclusion that the future of search needs to be in glowing neon lights. The conceptual jump from searches “rendering a list of links” to that final summary paragraph of requested information should be every search engine’s goal.

]]> By: CJ http://www.scienceforseo.com/information-retrieval/assisted-search/comment-page-1/#comment-1132 CJ Wed, 22 Jul 2009 01:51:54 +0000 http://www.scienceforseo.com/?p=1173#comment-1132 Hi Lucas, no it's a European company. The paper I wrote on user interaction with conversational systems covers everything I extracted from the user research. I looked at absolutely everything :) What customers search for is not always what they mean to look for, indeed. Using customer service logs, especially if the company does this online, means that there bis a wealth of data to work with. I'm not interested in predicting the user path right now, I'm only interested in language and generating language automatically really. KIA2 will be available at some point in the next months. I will let you know as soon as it is :)<!-- Touched by JuLiA --> Hi Lucas,

no it’s a European company. The paper I wrote on user interaction with conversational systems covers everything I extracted from the user research. I looked at absolutely everything :)

What customers search for is not always what they mean to look for, indeed. Using customer service logs, especially if the company does this online, means that there bis a wealth of data to work with. I’m not interested in predicting the user path right now, I’m only interested in language and generating language automatically really.

KIA2 will be available at some point in the next months. I will let you know as soon as it is :)

]]> By: Lucas Ng http://www.scienceforseo.com/information-retrieval/assisted-search/comment-page-1/#comment-1131 Lucas Ng Wed, 22 Jul 2009 01:27:44 +0000 http://www.scienceforseo.com/?p=1173#comment-1131 Sounds like you are doing some very exciting research CJ! "I also analyse a lot (seriously, it was overwhelming!) of customer service logs and the comparison between search es on the site and actual customer motivation was vastly different" Can you expand on this? Are you referring intent vs. query formulation? i.e. What the customer was searching for did not match what that customer typed? I also agree that user personas as typically used by User Experience architects are not terribly useful, however, for machine learning to occur on a commercial website, there would have to be an initial learning set and that has been hard for us to business case. Was the commercial website Australian? I work at Australia's largest media publisher and would be very interested in seeing your KIA system in action!<!-- Touched by JuLiA --> Sounds like you are doing some very exciting research CJ!

“I also analyse a lot (seriously, it was overwhelming!) of customer service logs and the comparison between search es on the site and actual customer motivation was vastly different”

Can you expand on this? Are you referring intent vs. query formulation? i.e. What the customer was searching for did not match what that customer typed?

I also agree that user personas as typically used by User Experience architects are not terribly useful, however, for machine learning to occur on a commercial website, there would have to be an initial learning set and that has been hard for us to business case.

Was the commercial website Australian? I work at Australia’s largest media publisher and would be very interested in seeing your KIA system in action!

]]>