问题
I'm new to the subject of NLP and requested to perform -named entity recognition- (NER) using Mallet. I have a text, and I give feature vector for each word in it. I would like to train a model which later on I can test on fresh text file. My question is how do I create such model, what is the input for the model. I could use some code examples :) Thanks !
回答1:
Fei Xia at UW wrote a pretty good MALLET guide.
You can find an example of programmatic (Java) interaction with MALLET at the bottom of this page.
回答2:
The MALLET quick start on sequence tagging, right on the MALLET home page, describes exactly how to do this if you already have the feature vectors.
来源:https://stackoverflow.com/questions/6776346/how-to-use-mallet-for-ner