Accurately analyzing and modeling online browsing behavior plays a key role in understanding users and technology interactions. Specifically, understanding whether users have correct perceptions of their browsing behavior will help to identify key features for models of user behavior, which will, in turn, enable realistic-looking synthetic data generation. In this work, we designed and conducted a user experiment to collect browsing behavior data from 32 participants continuously for 14 days. The collected dataset includes URLs of visited websites, actions taken on each website (such as clicking links or typing in a textbox), and timestamps of all activities. Finally, we use this new dataset to empirically address the following questions: (1) Do people have correct perceptions of their level of online behavior? (2) Do people alter their browsing behavior knowing that they are being tracked? (3) How do structural properties of browsing patterns vary across demographic groups?
- Tags
-