I am not actually into statistics but in computer science. I want to use log-likelihood in order to be able to detect which keywords are characteristics of a textual document based on frequency of occurence. I will need to compare the frequency in one text collections with the frequency of the same word in another textual collection.
I understand the basic mechanism of log-likelihood however I did not understand how to calculate the p-value along with the interpretation of this value.
Many thanks for your help,
Advertise on Talk Stats