To run the filtering on acl dataset, download the anthology from here to the datasets folder. Select the bibtex with abstracts version.
Then, run the filter_acl.py as
filter_acl.py > nlp_abstracts_for_keywords
The above will dump all the abstracts matching criteria defined in the filter_acl file. Edit the criteria as per your needs in the filter_acl script.