You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, I have read the paper and the released code. After that, I have some question about user node in tweet-user subgraph:
how did u get the representation of each user. You wrote in the paper, "the representation xui of each user ui can be extracted from the user behaviors or user profile data". So, what information does user profile data include? how did u extract from those user profile data?
Is the user feature dimension the same as the word feature dimension?I don't find the Equation 4(a process of projection to transform user feature and tweet feature) in ur released code.
In ur ind.twitter15.user.twitter.adj, the number of node is only 7085. I don't know the reason why the number is so small.
Looking forward to your reply, thank you very much.
The text was updated successfully, but these errors were encountered:
Hello, I have read the paper and the released code. After that, I have some question about user node in tweet-user subgraph:
Looking forward to your reply, thank you very much.
The text was updated successfully, but these errors were encountered: