Filter Bubbles and Content Diversity? An Agent-Based Modeling Approach


Personalisation algorithms play an important role in catering the information that is relevant to us. The best results are achieved by the algorithms when they monitor the user activity. Most of the algorithms adapt to the users’ personal preferences by filtering out the information that is irrelevant to the user. However, one of the criticisms of this process is that it is leading to informational bubbles called the filter bubbles which is a personal space of content familiar to the user, which would reinforce their confirmational biases or create informational blind spots. This phenomena however is highly debated. In this light, we propose an agent based model study, which tries to verify the implications claimed by the filter bubble theorists and also create an hypothetical environment that does not have a filter bubble and test difference in the information dispersion and opinion formation in both the environments.

In International Conference on Human-Computer Interaction