Overview of the dataset
We will be using a SNAP public dataset, social circles: Facebook, from Stanford University (https://snap.stanford.edu/data/ego-Facebook.html).
The dataset was created by collecting Facebook user information from survey participants. In more detail, ego networks were created for 10 users. Each user was asked to identify all the circles (list of friends) to which their friends belong. On average, each user identified 19 circles in their ego networks, where each circle has on average 22 friends.
For each user, the following information was collected:
- Edges: An edge exists if two users are friends on Facebook.
- Node features: Features are scored as 1 if the user has this property in their profile and 0 otherwise. Features have been anonymized since the names of the features would reveal private data.
The 10 ego networks were then unified in a single graph that we are going to study.
Dataset download
The dataset can be retrieved...
