Automatically assign semantics to large data sets from heterogeneous sources based on their features using several Statistical and Machine Learning techniques.
cd container; docker build -t isi/semantic-labeling .
docker-compose up
bin/semantic_labeling.sh <train_dataset> <test_dataset> <train_dataset2>