minghui/Twitter-LDA

What's the parameter "gamma" and "beta_word" stands for

Opened this issue · 2 comments

What's the parameter "gamma" and "beta_word" stands for

I was just start learning LDA and want to analyze the bullets from bilibili.com. But when I compare LDA and Twitter-LDA , Twitter-LDA has two more parameters ,"gamma" and "beta_word".After I read the thesis you had list in README. Knowing the model for T-LDA, I still can not find the explaination for those two parameters.So I ask for help if I missed something .
Thanks a lot

Please find them in Fig. 2 of the T-LDA paper: https://pdfs.semanticscholar.org/2cf3/79819632deb93b2cd9250da25bf21fa25171.pdf
beta_word and beta_background are priors for phi' and phi^B respectively, gamma is the prior for pi.