Utilizing Structural Network Positions to Diversify People Recommendations on Twitter

Table 1

Overview of typical recommendation approaches.

Approach	Type of recommended users
Similarity [41]	Users who share interests (tweet content) or audience (network)
Triadic closure [6, 19]	If A & B both follow C, recommend A to B (or vice versa); if A follows both B & C, recommend B to C; if A follows B and B follows C, recommend C to A
Popularity [42, 43]	Users who have many followers or followees
Activity [43]	Users who frequently tweet, mention, retweet, etc.
Reciprocity [44]	One’s followers
Activity & followership [44]	Users whose tweets are retweeted by followees of the target user