At the moment, I am working on a project which requires me to build a naive Bayes classifier. Right now, I have a form online asking for people to submit a sentence and the subject of the sentence, in order to build a classifier to identify the subject of a sentence. But before I train the classifier, I intend on processing all entries for the parts-of-speech and the location of the subject. So my training set will be formatted as:

Sentence: Jake moved the chair.   Subject: Jake
POS-Tagged: NNP VBD DD NN         Location: 0

Would this be an effective way to build the classifier, or is there a better method?


1 Answer 1


Your approach would definitely work. I would recommend training a variety of classifiers and comparing their performance using multiclass roc analysis. Also, think about other useful features in addition to the ones you mentioned (e.g. pos tag). Feature engineering is one of the most important factors in building good predictive models. Another thing to keep in mind is that the classes could be highly imbalanced which might influence your model's performance.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .