jkkummerfeld / berkeley-coreference-analyser Goto Github PK

Hi, 

  I am trying to use coreference_reading script for parsing coreference annotated file in CoNLL format. I would like to know what the function read_conll_coref() returns. Can you please help me with this ? If there is some documentation available of the whole system it would be great.

Thanks,
Joe

(5, 20, 23)    Extra:  the Giant Buddha
(17, 6, 14)    this world 's largest outdoor seated bronze Buddha
(22, 1, 5)     this giant bronze Buddha

Missing:
(19, 2, 9)     Hong Kong 's Tian Tan Giant Buddha

Raw changes:
merge 2
introduce 1
split 2
remove 1

Categorised:
1 Conflated Entities
1 Missing Mention

Detailed error listing:
split (set([(17, 6, 14), (22, 1, 5)]), set([(17, 6, 14), (22, 1, 5), (5, 20, 23)]), '', ['split', 2, 1, None, '0_cataphoric', 2, 0, 0, 1, 0, 0, False, True, 'no_string_match', 'head_match', 'merge', 'split', False, 'part_', 'cluster_WORK_OF_ART', True, 'part_', 'cluster_', True, 'part_', 'cluster_', True, 'part_', 'cluster_'])
Properties included: split ['split', 2, 1, None, '0_cataphoric', 2, 0, 0, 1, 0, 0, False, True, 'no_string_match', 'head_match', 'merge', 'split', False, 'part_', 'cluster_WORK_OF_ART', True, 'part_', 'cluster_', True, 'part_', 'cluster_', True, 'part_', 'cluster_']
missing mention (19, 2, 9)     Hong Kong 's Tian Tan Giant Buddha
missing mention (set([(19, 2, 9)]), set([(17, 6, 14), (22, 1, 5), (19, 2, 9)]), (set([(19, 2, 9)]), set([(17, 6, 14), (22, 1, 5), (19, 2, 9)]), ['merge', 1, 2, "hong_kong_'s_tian_tan_giant_buddha", '0_cataphoric', 1, 0, 0, 2, 0, 0, False, False, 'no_string_match', 'head_match', 'merge', 'introduce', True, 'part_', 'cluster_', True, 'part_', 'cluster_', True, 'part_', 'cluster_', True, 'part_', 'cluster_']), ['missing', 'name', "hong_kong_'s_tian_tan_giant_buddha", 'no_text_match', 'head_match', 'not_nested', False, False, False, 'ner_unknown', 'number_unknown', 'person_unknown', 'gender_unknown'])
Properties included: missing mention ['missing', 'name', "hong_kong_'s_tian_tan_giant_buddha", 'no_text_match', 'head_match', 'not_nested', False, False, False, 'ner_unknown', 'number_unknown', 'person_unknown', 'gender_unknown']