Online market creates too many choices

Many of us turn to personal recommendations when deciding to make a purchase, but the sheer volume of online customer feedback now available can be just as daunting.

Leading online retailers – notably Amazon – were quick to develop personal recommendation software, analysing a customer’s buying patterns to suggest other products that they might enjoy.

A team of European researchers are now taking that approach a step further.

Alexander Voss, a researcher at Microsoft’s European Innovation Centre in Aachen, Germany, is coordinating seven companies and universities that are developing advanced methods, models and algorithms to bring personalised recommendations to everything from web content and interactive TV channel guides to e-commerce.

Their algorithms, developed in the EU-funded MyMedia project, combine two sources of data. “Stable information” is derived from user-created profiles – covering details such as age, residence, likes and dislikes – and the star∞ratings they give to past purchases, such as films they have enjoyed.

“Unstable information” reflects factors such as context, the user’s mood and tastes, and what types of music they tend to listen to.

Most personalised recommendation technology looks at unstable information as a black and white issue: if a user watched one video they must want to see more like it.

But the MyMedia researchers are adding a third and very important variable when it comes to finding what people really want. “We’re not just looking at choices in terms of ‘yes’ or ‘no’, but also ‘maybe’,” Voss explains.

“Just because someone didn’t watch a certain video doesn’t mean they wouldn’t want to, they simply might not have had the time or felt like it at that moment.

“Instead of excluding it, we add it to a relative ranking of recommended content that changes over time as the system builds up a better idea of the user’s interests.”

The system is unique in another way too: it allows content providers to fine-tune recommendation parameters to see how the changes affect users’ responses. “There is no other recommender system that can do that,” Voss says.

The technology is being tested by e-commerce company Microgénesis, telecoms company BT and the BBC. If it catches on, we could soon have more time to spend enjoying our choices, rather than worrying about what choice to make.