I need to classify the text docs from elasticsearch using naive bayes classifier. I experimented on nltk but it doent have support for incremental or stream data handling. I referred to the below doc
Is it possible to do incremental training with H2O if yes, how? i am also open to use some other classifier which supports incremental mining.
H2O-3 has the option for checkpointing, though not for Naive Bayes. from the docs:
The checkpoint option is available for DRF, GBM, and Deep Learning algorithms. This allows you to specify a model key associated with a previously trained model. This will build a new model as a continuation of a previously generated model. If this is not specified, then the algorithm will start training a new model instead of continuing building a previous model.
If this is what you are looking for, the above link also links to python and R code examples on how to use the checkpointing parameter.
