CIOTechie_ArtificialIntelligence_DeepMind_AI_Google_Alphabet,Inc._Recommender-System_Analysis_User_Behavior

Researchers At DeepMind Offer Theoretical Analysis To Analyze Users’ Behavior On The Recommender System

Artificial Intelligence News

Researchers-At-DeepMind-Offer-Theoretical-Analysis-To-Analyze-Users’-Behavior-On-The-Recommender-System Researchers At DeepMind Offer Theoretical Analysis To Analyze Users’ Behavior On The Recommender SystemSearch engine giant Google’s DeepMind researchers have issued a paper last week, to provide a new theoretical analysis to examine the user dynamics role and the behavior of recommender systems. The paper, titled ‘Degenerate Feedback Loops in Recommender Systems’, can assist them to remove the echo chamber from the filter bubble effect.

Designed to provide users with personalized product and information offerings, Recommender systems take into consideration the user’s personal characteristics and past behaviors to make a list of items that have been personalized as per the user’s interests. There are certain concerns related to the systems that it might lead to a self-reinforcing pattern of narrowing exposure and a shift in user’s interest, and these issues are often called the echo chamber and filter bubble. On the echo chamber, the researchers, in the paper, explained that user’s interest being optimistically or negatively reinforced owing to the repeated exposure to a certain category of items. And for filter bubble, they noted that the recommender systems select limited content to serve the users online.

Furthermore, they have considered that a recommender system which is able to interact with a user over time; at every time step, it serves a diverse number of items or categories of items like news articles, videos, or consumer products, to a user from a set of finite or countable infinite items. With the paper, the researchers also considered the fact that the user’s interaction with the recommender system can change based on their interest in various items for the next interaction. Moreover, to further evaluate the echo chamber or filter bubble effect in recommender systems, researchers track when the user’s interest changes awfully.