Improving human interaction with artificial advice givers

Themes: Software Technology & Intelligent Systems

A TRL is a measure to indicate the matureness of a developing technology. When an innovative idea is discovered it is often not directly suitable for application. Usually such novel idea is subjected to further experimentation, testing and prototyping before it can be implemented. The image below shows how to read TRL’s to categorise the innovative ideas.

Click to enlarge

Summary of the project

The researcher works on generating explanations: understanding and developing methods for recommender systems, such as Amazon or Spotify, to become more transparent. Based on your choices for a book or music you get suggestions of other books or music that you will like, or is liked or bought by others who have made similar choices in the past. These suggestions are generated through an algorithm which, based on the preference data of users of such a platform, will make a prediction of what else you would like.

The researcher has a unique approach which combines studies of algorithms on the one hand and is user centred on the other hand. On the algorithm side the researcher looks at the content selection and how this can be improved to consider not only relevance but also other criteria such as diversity. From a user centered perspective the researcher looks at how much influence people should have on the system, and how this differs for different user properties (e.g., experts, cognitive capacity), or usage contexts (e.g., focused or relaxed). Currently the researcher is looking at personalised news platforms and how to use these methods to ensure that users are aware of a wider range of views.

What's next?

The next step is to look at how to explain recommendations to a group of people, where people have different preferences. Future plans also include improving ways of measuring viewpoint diversity in online conversations on Twitter.

Dr. Nava Tintarev


Shabnam Najafian

Dr. Dimitrios Bountouridis

Dr. Emily Sullivan

Dr. Mark Alfano

Yucheng Jin

Dr. Katrien Verbert