cltl / multilingual_factuality

Multilingual factuality

This repository is used to build a multi-lingual system for identifying the factuality of events. We are currently working on a basic implementation for Dutch and English.

Running the module

The scripts to run the module are located at feature_extractor/

1. On a single file:

cat inputfile | python rule_based_factuality.py > outputfile

2. For a directory:

./run_rule_based_on_dir.sh inputdir/ outputdir/

Version:

Version 0.01 is released on May 9th 2016. It is a basic implementation and first creation of resources based on intuition. We expect to have many minor revisions in the near future. You will therefore find some version 0.xx in the document.

As soon as the resources for at least one language and the rule application system have stabilized, we will release version 1.0

Content

This repository consists of:

docs/ for documentation data/ for development and training data resources/ for language specific resources (vocabularies) and models scripts/ for code extracting input features, applying rules/calling machine learning modules and producing output.

Evaluation/experimental setup:

please check out: https://github.com/cltl/factuality_experimental_environment

For gold data, input data and evaluation scripts for factuality.

Contact

Antske Fokkens: [email protected] Ruben Izquierdo: [email protected] Roser Morante: [email protected] Tommaso Caselli: [email protected]

multilingual_factuality's Issues

KeyError: t_xxx

I seem to get this error occasionally when running the multilingual factuality.

Traceback (most recent call last):
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 413, in <module>
    main()
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 406, in main
    run_factuality_module(nafobj)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 392, in run_factuality_module
    events_features = extract_features(feature_extractor, target_events)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 367, in extract_features
    add_predicate_chain_features(feature_extractor, event, myFeatures)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 210, in add_predicate_chain_features
    pred_chain = feature_extractor.get_list_term_ids_to_root(tid)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/my_feature_extractor.py", line 173, in get_list_term_ids_to_root
    root_for_sentence = this_graph.get_root()
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/my_feature_extractor.py", line 41, in get_root
    self.calculate_root()
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/my_feature_extractor.py", line 35, in calculate_root
    list_with_min_freq = [(term_id, len(self.G[term_id])) for term_id, freq in L if freq == min_freq]
KeyError: 't_840'

An example input file that causes the error can be found here: http://i.amcat.nl/keyerror.naf

TypeError: 'NoneType' object is not iterable

I get this error when running the factuality

Traceback (most recent call last):
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 413, in <module>
    main()
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 406, in main
    run_factuality_module(nafobj)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 392, in run_factuality_module
    events_features = extract_features(feature_extractor, target_events)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 367, in extract_features
    add_predicate_chain_features(feature_extractor, event, myFeatures)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/rule_based_factuality.py", line 212, in add_predicate_chain_features
    eventObj.predicate_chain_lemmas = feature_extractor.get_lemmas_for_list_term_ids(pred_chain)
  File "/data/wva/newsreader_pipe_nl/modules/multilingual_factuality/feature_extractor/my_feature_extractor.py", line 182, in get_lemmas_for_list_term_ids
    for term_id in list_term_ids:
TypeError: 'NoneType' object is not iterable
WARNING:elasticsearch:HEAD /nlpipe/annotate__0_0 [status:404 request:0.001s]

As far as I can see, get_list_term_ids_to_root returns None in some cases, which is not expected by the get_lemmas_for_list_term_ids. Adding or None to the return value of the former function prevents the error, but I have no clue what I'm doing so that's probably not a solution.

Let me know if you need the raw input that causes the problem.