4

I want to use the majority of the out-of-the-box classifier than TextBlob offers, but I also wanted to add my own small set of training data. This is because the text I am analyzing has some niche words I want to make sure make it into the training set.

So, in TextBlob, they say you can augment an existing classifier like this

>>> new_data = [('She is my best friend.', 'pos'),
        ("I'm happy to have a new friend.", 'pos'),
        ("Stay thirsty, my friend.", 'pos'),
        ("He ain't from around here.", 'neg')]
>>> cl.update(new_data)
True
>>> cl.accuracy(test)
1.0

However, it doesn't say anything about adding this data to the default classifier. Does anyone know if this is possible?

EDIT

Alternatively, is there a place where I can get enough training data so that I can training my classifier the other way around?

0 Answers0