Out-degree counts the outgoing relationships connected to a node in a directed network. In a follower network, it represents how many users each user follows. section 5.1.1
Use out-degree to inspect network structure and compare nodes:
- Separate it from in-degree, which counts incoming relationships. section 5.1.1
- Summarize its distribution with minimum, maximum, mean, standard deviation, and percentile values. section 5.1.1
- Interpret a high out-degree as many outgoing connections, while a low or zero value indicates few or no outgoing connections. section 5.1.1
- Use weighted out-degree when relationship weights represent repeated or stronger connections. section 6.2.1
Listing 5.1 Evaluating the node out-degree distribution with the apoc.agg.statistics function
MATCH (u:User)
WITH u, count{ (u)-[:FOLLOWS]->() } AS outDegree
RETURN apoc.agg.statistics(outDegree)
Calculate each user's out-degree in Cypher, then summarize the resulting values with APOC statistics. section 5.1.1
Follower-network distribution
A followers network with 3,594 users has a mean out-degree of about 6.9, a median of 2, and a 90th percentile of 21. Thus, half of the users follow two or fewer users, while about 10% follow more than 21. section 5.1.1
Listing 6.4 Evaluating the out-degree distribution of the inferred retweet amplification network
CALL gds.degree.stats('amplify')
YIELD centralityDistribution
Evaluate out-degree statistics on an inferred in-memory retweet amplification network with the GDS degree algorithm. section 6.2.1
