Microsoft makes Bing search more personal

Microsoft has altered its Bing search pages in a bid to make them more relevant and useful to individual users.

Search is a hotly contested area, and just this month Microsoft was accused of copying Google's results.

Hoping to stake more of a claim on the market, Microsoft is planning to learn from its users' habits and provide an experience based on their history, preferences and acknowledged interests.

If Microsoft cracks it, and makes its search results quicker and more accurate, it could find favour among mobile workers and time-pressed enterprise users.

The updates are the result of much experimentation and investment in search improvements, which the firm claimed has elevated Bing above generic predictive results.

Microsoft explained that the biggest obstacle facing search engineers is the simple fact that human behaviour is not predictable.

"Don't get us wrong; people can be creatures of habit and we can build functions that enable us to display different results based on logical assumptions made in the aggregate. But what if we're wrong?" said Aidan Crook and Sanaz Ahari of the Bing Search team.

One of the challenges in delivering individualised results is that personalised search "can't see the forest for the trees", according to Microsoft.

"In other words, everyone is collecting everything and trying to figure out the foibles of human behaviour from a mass of digital bits," said Crook and Ahari.

Rather than use just data built over a period of time from search histories, Microsoft will take other real-time information into consideration.

"In real life, you take into account many things when making a decision: for example, where you are, what time it is, and what your friends think. Shouldn't your search engine do the same thing?" wrote the engineers.

In an attempt to replicate this, Microsoft is experimenting with a number of techniques, including the automatic tailoring of search results based on a person's physical location and past searches.

"The beauty of thinking differently about personalised searching is that it enables us to construct elegant solutions that require a minimal amount of personal information and, frankly, often exhibit better results than a more computationally complex predictive model alone," added Crook and Ahari.

The feature is currently limited to the US, but will be rolled out globally at some point.