Evaluation: Relative Cross Entropy (RCE)

fani-lab / SEERa

A framework to predict the future user communities in a text streaming social network based on the users’ topics of interest.

Other

4 stars 5 forks source link

Open soroush-ziaeinejad opened 2 years ago

soroush-ziaeinejad commented 2 years ago

This metric measures how much better a predictor is compared to a naive prediction. It seems so fit for our news article recommendation application.

RCE = 100 × (Reference Cross Entropy − Cross Entropy) / Reference Cross Entropy

hosseinfani commented 2 years ago

Hi @farinamhz Are you sure you're putting your summary in the correct issue page? I see no relation between lda and rce?

farinamhz commented 2 years ago

Hello @hosseinfani, I am sorry, I posted my summary in the wrong place. I fixed it.