Community-Based Recommendations on Twitter: Avoiding the Filter Bubble

摘要: Due to their success, social network platforms are considered today as a major communication mean. In order increase user engagement, they rely on recommender systems personalize individual experience by filtering messages according interest and/or neighborhood. However some recent results exhibit that this personalization of content might the echo chamber effect and create filter bubbles. These bubbles restrain diversity opinions regarding recommended content. paper, we first realize thorough study communities large Twitter dataset quantify how affect users’ behavior Then propose Community Aware Model (CAM) counter impact different information consumption. Our show concern up 10% users our model based similarities between enhance systems.