Thursday, 26 March 2009

Google tests more text with each snippet

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 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.

Friday, 6 March 2009 been giving term suggestions?

How long has been providing this term suggestion on its interface? with the number of results its going to provide? They don't do it on 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 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.