Daniel Tunkelang has brought my attention to another blog entry about some of the tests that Google is carrying out at the moment. As well as letting you view timelines, and a 'wonder wheel' of connections, the options it lets you test include adding thumbnails to each search result (something that Ask.com has been doing for a while) and also allowing you to see more than 2 lines of text per result.
This last point is the one that seems rather interesting to me. I've heard many a search engine representative talk about getting as many results as possible above the fold (the point where you'd have to scroll to keep reading), and getting the best trade off, therefore, between context and space. Tim Paek et al, at Microsoft Research, studied the idea of flexible snippet lengths back at CHI2004. Its been a long time coming. I proposed back at a SIGIR workshop in 2007 that we just let people choose the size of the each snippet in the preferences, and see how often people change it - and to what? Maybe now we'll see.
Interestingly, in IEEE Computer in March 2009, Daniel Russell, of Google, wrote an article saying that, for some research, only big corporations with thousands of processors and millions of users can really test small UI changes, among many other things. Well I'm glad that Google is testing this - and I hope we see some results from it too.
Thursday, 26 March 2009
Friday, 6 March 2009
google.com been giving term suggestions?
How long has google.com been providing this term suggestion on its interface? with the number of results its going to provide? They don't do it on google.co.uk. Fun fun.
Thursday, 5 March 2009
What seperates query refinement, clustering, and faceted search?
I've been thinking recently about what seperates out the different interactive information retrieval techniques, as a term I am using loosely for now. There's interactive query refinement or expansion, which is often used to suggests potential changes to a query to explore sub-groups of the results. There's clustering, which analyses the results for clusters, in order to help users explore sub-groups of the results. And there's faceted search, which provides many different types of categorisation over the results in order to help users explore sub-groups of the results.
Each of these can be used to explore groups in the results, and they mainly differ by the back-end system that is used to label the sub-groups. They each also come with a typical interaction model. IQE usually sends a new query to the server and returns a new set of results. Clustering interaces, like Clusty.com typically allow users to choose one cluster at a time to view. Faceted browsers, like Flamenco or mSpace, typically allow users to apply and unapply a series of filters.
My question is how much of the effect is down to the method and which is down to the interaction model. Marti Hearst wrote a great article in the CACM that highlighted the advantages of faceted exploration over clustering, but the majority of her highlights are over quality of data produced, such as the completeness of categories produced.
It would be interesting to compare the specific effect of interaction style. Such as allowing users to apply and unapply a series of interactive query refinemements, rather than sending off new queries as a new starting point. The nearest I can think to research doing this is the work by Hoeber, which allows users to turn on and off query refinement filters on the list of results. The aim of such a specific study would be to analyse the benefit of implementing more increasingly complicated backends, instead of simply improving the interactivity of the search interface and the range of search tactics they support.
Each of these can be used to explore groups in the results, and they mainly differ by the back-end system that is used to label the sub-groups. They each also come with a typical interaction model. IQE usually sends a new query to the server and returns a new set of results. Clustering interaces, like Clusty.com typically allow users to choose one cluster at a time to view. Faceted browsers, like Flamenco or mSpace, typically allow users to apply and unapply a series of filters.
My question is how much of the effect is down to the method and which is down to the interaction model. Marti Hearst wrote a great article in the CACM that highlighted the advantages of faceted exploration over clustering, but the majority of her highlights are over quality of data produced, such as the completeness of categories produced.
It would be interesting to compare the specific effect of interaction style. Such as allowing users to apply and unapply a series of interactive query refinemements, rather than sending off new queries as a new starting point. The nearest I can think to research doing this is the work by Hoeber, which allows users to turn on and off query refinement filters on the list of results. The aim of such a specific study would be to analyse the benefit of implementing more increasingly complicated backends, instead of simply improving the interactivity of the search interface and the range of search tactics they support.
Friday, 27 February 2009
Concert for the deaf?
One of the most amazing people I ever had the pleasure of working with, is putting on a multi-sensory concert for the deaf. Her work on modelling the human cochlea is being tested as part of a audio-responsive chair in a live concert designed for the hearing impaired.
I'm sure it will be an amazing experience for both the gig-goers, the bands, and the researchers seeing their creative work in action!
I'm sure it will be an amazing experience for both the gig-goers, the bands, and the researchers seeing their creative work in action!
Thursday, 26 February 2009
Is Web-based Exploratory search on the increase?
I read an interesting paper by Vakkari's team, on the different queries submitted to libraries via an online form, between 1999 and 2006. The trends are quite interesting, and one of the conclusions is that topic-related searches have reduced in libraries because they are, instead, being performed more on the web. This creates two questions about topic-searching on the web:
1) many HCIR style papers assume that this is hard to do on the web, but this research suggests its happening more anyway. This is perhaps because its more convenient to access the web now, than it is to drive across town. The service they analysed, however, was an online library query service (in Finland).
2) This is surely motivation for providing better exploratory search interfaces on the web, to help people explore and learn topics - why has it only dropped from 57% to 47%? Why not further?
They also conclude that people still turn to librarians for difficult searching problems. This really is motivation for providing better exploratory search interfaces, so that a) the number of topical searches to libraries goes down even more and b) so that the number of difficult questions goes down instead of up!
1) many HCIR style papers assume that this is hard to do on the web, but this research suggests its happening more anyway. This is perhaps because its more convenient to access the web now, than it is to drive across town. The service they analysed, however, was an online library query service (in Finland).
2) This is surely motivation for providing better exploratory search interfaces on the web, to help people explore and learn topics - why has it only dropped from 57% to 47%? Why not further?
They also conclude that people still turn to librarians for difficult searching problems. This really is motivation for providing better exploratory search interfaces, so that a) the number of topical searches to libraries goes down even more and b) so that the number of difficult questions goes down instead of up!
Monday, 16 February 2009
search interaction is short
*warning - read the comments below before you read the article discussed here*
I came across an interesting article which is, to some extent, both a challenge for interactive information retrieval, and a blow to idea that search should be like a conversation (rather than guessing a searcher's intentions). One of their notable findings is that the average search session is 2.9 interactions long. Nice to see that its not considering search session length in terms of time (a common metric, but not always applicable during information seeking), but instead in the number of interactions. This is something in the vein of my own research.
This finding really only allows 1) an initial search, 2) an interactive refinement and/or a scroll, and 3) a selection. This also assumes, since the 2.9 is less than 3, that one of these is optional. and its unlikely to be the searching or the selecting. I want to go over the paper in some more detail, but its certainly interesting.
I came across an interesting article which is, to some extent, both a challenge for interactive information retrieval, and a blow to idea that search should be like a conversation (rather than guessing a searcher's intentions). One of their notable findings is that the average search session is 2.9 interactions long. Nice to see that its not considering search session length in terms of time (a common metric, but not always applicable during information seeking), but instead in the number of interactions. This is something in the vein of my own research.
This finding really only allows 1) an initial search, 2) an interactive refinement and/or a scroll, and 3) a selection. This also assumes, since the 2.9 is less than 3, that one of these is optional. and its unlikely to be the searching or the selecting. I want to go over the paper in some more detail, but its certainly interesting.
ambiguous query terms
Since my last entry, on what to do with more generic query terms, I have come across a few sources about this. First, I happened to review a paper on the topic, which I of course can't say more about. Second, I have happened upon an interesting journal article looking at identifying ambiguous terms. It's by no means the only research to try and do this, but their recent work has found only around 16% of online queries are what they define as ambiguous.
Finally, an interesting blogger, has mentioned an alternative search engine called DuckDuckGo, which, I'm pleased to say, does almost exactly as I discussed in my previous entry. As you can see with the standard ambiguous example of apple, it breaks down results into groups that cover a range of its different domain relations, which can be used for interactive query expansion. Give it a try. They have a nice list of their defining features. I'm currently using it as my default search engine now too.
Finally, an interesting blogger, has mentioned an alternative search engine called DuckDuckGo, which, I'm pleased to say, does almost exactly as I discussed in my previous entry. As you can see with the standard ambiguous example of apple, it breaks down results into groups that cover a range of its different domain relations, which can be used for interactive query expansion. Give it a try. They have a nice list of their defining features. I'm currently using it as my default search engine now too.
Subscribe to:
Posts (Atom)
