issues
search
flaprimo
/
twitter-network-analysis
Directed network analysis, for a twitter data source
8
stars
2
forks
source link
Question on Further Work
#12
Open
BradKML
opened
3 years ago
BradKML
commented
3 years ago
Is it useful to do role detection (e.g. RolX, RoleSim) and niche centrality measures (NetworkX) to discover persons of interest?
Would alternate Community Detection (CDLib) algorithms be better at network analysis?
Would Node Embedding (KarateClub) be useful at simplifying the structuring of the network?
What are the major indicating factors of Temporal Directed Networks?
What is the best way to weigh a user network, where each user pair can only have one edge but multiple retweets and replies?
Are there any differences between these three items?
conversation community (retweet/reply/citations networks)
belief community (follow networks)
tweet topic models (e.g. TF-IDF with Tweet Keywords and Hashtags)
BradKML
commented
3 years ago
Other questions from other projects:
Possibility of quantifying Polarization (or Multiplicity)
https://github.com/tedhchen/twitter_network_polarization_tools