Knowledge base creation app for family relations

#Pipeline for preprocessing:

1. Filter raw warc data by document id and copy to database: ./setup_database.sh This will copy the filtered sentences to table sentences_intermediate_warc

2. Sample sentences randomly to create a development data set: First, change the last LIMIT statement in sample_sentences.sh to the number of documents you want, or exclude the limit statement if you want all sentences.

Then run: ./sample_sentences.sh

This will insert the randomly sampled data to table sentences