I want to train text classifier (using https://www.uclassify.com) with 12 classes/categories. I will be training it to classify news/articles (I know that there are existing classifier but I want to train my own).
uclassify uses following algorithm (directly copied from their site):
The core is a multinominal Naive Bayesian classifier with a couple of steps that improves the classification further (hybrid complementary NB, class normalization and special smoothing). The result of classifications are probabilities [0-1] of a document belonging to each class. This is very useful if you want to set a threshold for classifications. E.g. all classifications over 90% is considered spam. Using this model also makes it very scalable in terms of CPU time for classification/training.
I was wondering how many examples I will need to train such classifier? It is possible to estimate the number? Let's assume that one article will "fit" 2 categories by average.