Adaptive search – personalised search evolved

by Peter Young on September 14, 2011 · 0 comments

Search engines have a very difficult job having to deal with a multitude of issues – however possibly one of the more complex is intent. Without context many search engine phrases typed into search engines may relate to a multitude of different things and without further qualification a search engine may struggle to return relative your particular requirements.

Bing have already made some early inroads into integrating personalisation into their results, with the development of a localised resultset and then considering previously visited sites into the result framework. At SMX today, Bing announced adaptive search.

So how does adaptive search work. Bing describe this as follows

Every time you search on Bing, the information provided helps Bing understand what you’re trying to do. The more you search, the more Bing can learn – and use that information to adapt the experience so you can spend less time searching and accomplish what you set out to do.

As an example, let’s say you’re in the process of planning a vacation – you might decide to search for “Australia”. In this case, you’re most likely to be looking for websites specifically about the country Australia, or information about travel. You can see an example of the search “Australia”, for someone who is planning a vacation, below:


Source: Bing

Now suppose, instead, you’re a movie-buff and are trying to decide on a movie to rent for the evening. With this context, the smart technology powering this feature will infer that you’re probably looking for the movie “Australia”, and begin to adapt the search page to your intent by showings results relevant to the movie Australia higher up on the page than they were previously:


Source:Bing

The differences are generally quite subtle, but the more confidence we have about what your intent is, the more personalized the results will become. We certainly don’t want to make any assumptions that prevent you from seeing a diverse set of results and lock you into a “filter bubble”, so the results that correspond to differing intents (e.g. travel to Australia) will still be available to you on the page.

Check out the following video to learn more about Adaptive Search:

<a href='http://video.msn.com/?mkt=en-us&#038;vid=60353988-525a-4bcf-92e7-df8cd4f57ce5&#038;src=SLPl:embed::uuids' target='_new' title='Adapting Search to You'>Video: Adapting Search to You</a>

It would appear these adaptive/personalised results are a lot easier than turning them in Google – and it would appear personalised search is something Bing would appear to have a significant eye on. The adaptive search results are not being rolled out beyond the US initially and timescales for more widespread rollout are still unclear.

I personally am liking a lot of the stuff Bing are doing at the moment, and I would suggest its not a surprise to see them taking market share off Google. The Facebook relationship as well should it develop is something I am positive Google in particular will be hoping doesn’t develop as there is no doubting Google see Facebook as a potentially significant competitor.

Theres no doubting Bing are improving – but are they improving fast enough.

Google+ Comments

Leave a Comment

Previous post:

Next post: