covid-texts

Run Clustering

$ python3 process.py -i {input-file} -o {output-file}

# e.g.
$ python3 process.py -i data/sample_input.json -o data/sample_output.json

Note: The input data structure should be like data/sample_input.json. And given that the algorithm does not perform well with short texts, we recommended to only keep items having more than 20 tokens in the input dataset.