A UCD scientist has made a major breakthrough in online “recommender” systems. Professor Barry Smyth has been granted US and China patents on his system to improve online shopping and information.
Recommender systems are now an important part of leading online stores including Amazon, iTunes and TripAdvisor. They help consumers to cope with a huge range of choices.
Professor Smyth says his new technique will help existing recommender systems to produce recommendations that are both relevant to users and different from each other.
What it does
Conventional recommender systems generate suggestions by picking items (e.g. movies, books, or music) that are like items that other similar users have enjoyed in the past.
For example, Amazon gives suggestions about what other similar users have recently purchased.
“One limitation is a tendency to produce sets of recommendations that are similar to each other,” Professor Smyth says.
Supposing a Harry Potter fan is given the top recommendations of “Harry Potter and the Chamber of Secrets” (Book 2), “Harry Potter and the Prisoner of Azkaban” (Book 3) and perhaps the latest DVD release for “Harry Potter and the Order of the Phoenix” (Book 5).
These recommendations are all very similar to each other, but if the user is not so interested in Harry Potter at this particular point in time, the recommender system is wasting valuable screen-space by making these near-duplicate suggestions.
A more effective approach involves selecting one Harry Potter suggestion and then complementing this with other recommendations that are also relevant to the user but that satisfy other tastes and preferences that they might have.
Professor Smyth’s new system describes how to do this, in a way that is compatible with existing recommendation techniques. The patents were filed by NovaUCD, the on-campus innovation centre, on behalf of UCD and Professor Smyth.
Major research field
Recommender systems are one of the most active research fields in computer science today. Irish researchers are at the forefront in this research. “Research groups in UCD and UCC are ranked among the leaders in recommender systems research,” says Professor Smyth.
Leading US online movie rental company Netflix even has a $1 million open competition to develop the next generation of recommender systems.
Learn more
Read about other research by Professor Smyth and his team at Clarity – the Centre for Sensor Web Technologies at UCD
Find out about how to take part in the Netflix competition
