Table 1 Feature list for TwiBot-20 dataset
From: Detecting bots in social-networks using node and structural embeddings
Type | Feature | Description |
|---|---|---|
Profile | protected | When true, indicates that this user has chosen to protect their Tweets. |
Profile | \(followers\_count\) | The number of followers this account currently has. |
Profile | \(friends\_count\) | The number of users this account was following (AKA their “followings”) at the time. |
Profile | \(listed\_count\) | The number of public lists that this user is a member of. |
Profile | \(favourites\_count\) | The number of Tweets this user has liked in the account’s lifetime. |
Profile | \(statuses\_count\) | The number of Tweets (including retweets) issued by the user. |
Profile | \(geo\_enabled\) | When true, indicates that this user enables geolocation (deprecated in today’s API). |
Profile | \(default\_profile\) | When true, indicates that the user has not altered the theme or background of their user profile. |
Profile | \(default\_profile\_image\) | When true, indicates that the user has not uploaded their own profile image and a default image is used instead. |
Profile | business | User interest domain: business (from [15]). |
Profile | entertainment | User interest domain: entertainment (from [15]). |
Profile | politics | User interest domain: politics (from [15]). |
Profile | sports | User interest domain: sports (from [15]). |
Profile | verified | True, if the user if verified by Twitter. This is the blue check-mark indicator in user’s profile. |
NLP | \(raw\_tweets\) | The raw recent 200 tweets for each root user. |
Graph | graph* | Degree Centrality, Strength, Eigen Centrality, Closeness, Harmonic Centrality, Betweenness, Authority, Hub Score, Constraint, Coreness, Eccentricity, PageRank |