alanyuchenhou / elephant

MIT License

4 stars 5 forks source link

communication prediction #27

Closed alanyuchenhou closed 7 years ago

alanyuchenhou commented 8 years ago

description

In a social network, communication frequency is a good indicator of tie: the more 2 users communicate (message, call, email), the stronger their potential tie is. This is shown in paper All Friends are Not Equal: Using Weights in Social Graphs to Improve Search. Predicting the potential communications between users enable us to recommend friends with stronger ties.

definition

graph:

node set: users
link set: communications (weight: amount of messages, calls, emails, etc.)

task: communication = predict(user1, user2)

dataset

Facebook-like Social Network

baseline solution and experiment

I cannot find any. Need help.

alanyuchenhou commented 7 years ago

This is succeeded by #34 :

problem: weight prediction
solution: weighted stochastic block model
dataset: forum