def __setup_entity_pairs_file():
# docs_ner_file = 'e:/dc/nyt-world-full/processed/docs.txt'
# ner_result_file = 'e:/dc/nyt-world-full/processed/ner-result.txt'
# cooccur_mentions_file = 'e:/dc/nyt-world-full/processed/mentions-ner/cooccur-mentions.txt'
# entity_name_dict_file = 'e:/dc/nyt-world-full/processed/mentions-ner/entity-names-nloc.txt'
# ner_result_file = 'e:/dc/nyt-world-full/processed/ner-result.txt'
# datadir = 'e:/data/emadr/nyt-world-full/processed/'
datadir = 'e:/data/emadr/nyt-less-docs/sports'
filter_loc = False
docs_ner_file = os.path.join(datadir, 'docs.txt')
ner_result_file = os.path.join(datadir, 'ner-result.txt')
cooccur_mentions_file = os.path.join(datadir, 'cooccur-mentions.txt')
entity_name_dict_file = os.path.join(datadir, 'entity-names.txt')
doc_all_mentions_file = os.path.join(datadir, 'doc-mentions.txt')
ee_file = os.path.join(datadir, 'bindata/ee.bin')
de_file = os.path.join(datadir, 'bindata/de.bin')
cnts_file = os.path.join(datadir, 'bindata/entity-cnts.bin')
dataarange.gen_ee_pairs_with_ner_result(docs_ner_file, ner_result_file,
cooccur_mentions_file)
# gen entity name dict
dataarange.gen_entity_name_dict(ner_result_file, entity_name_dict_file,
filter_loc)
dataarange.ner_result_to_tab_sep(ner_result_file, doc_all_mentions_file)
dataarange.gen_doc_entity_pairs(entity_name_dict_file,
doc_all_mentions_file, de_file)
dataarange.gen_entity_entity_pairs(entity_name_dict_file,
cooccur_mentions_file, ee_file)
dataarange.gen_cnts_file(de_file, cnts_file)
def retrieve_mentions():
line_docs_file_name = 'e:/dc/nyt-world-full/processed/docs-tokenized.txt'
illegal_start_words_file = 'e:/dc/20ng_bydate/stopwords.txt'
output_dir = 'e:/dc/nyt-world-full/processed/mentions/'
dst_doc_entity_candidates_list_file = output_dir + 'doc_mention_candidates.txt'
dst_entity_candidate_clique_file = output_dir + 'mention_candidate_cliques.txt'
dst_doc_entity_indices_file = output_dir + 'doc_mention_candidate_indices.txt'
# init_entity_net(line_docs_file_name, illegal_start_words_file, dst_doc_entity_candidates_list_file,
# dst_entity_candidate_clique_file, dst_doc_entity_indices_file)
lc_word_cnts_file_name = 'e:/dc/el/wiki/wiki_word_cnts_lc.txt'
wc_word_cnts_file_name = 'e:/dc/el/wiki/wiki_word_cnts_with_case.txt'
proper_entity_dict_file = output_dir + 'entity_names.txt'
# gen_entity_name_dict(dst_doc_entity_candidates_list_file, lc_word_cnts_file_name, wc_word_cnts_file_name,
# proper_entity_dict_file)
de_file_bin = 'e:/dc/nyt-world-full/processed/bin/de.bin'
de_file_txt = 'e:/dc/nyt-world-full/processed/bin/de.txt'
dataarange.gen_doc_entity_pairs(proper_entity_dict_file,
dst_doc_entity_candidates_list_file,
de_file_bin, de_file_txt)
entity_candidate_cliques_file = dst_entity_candidate_clique_file
ee_file = 'e:/dc/nyt-world-full/processed/bin/ee.bin'
dataarange.gen_entity_entity_pairs(proper_entity_dict_file,
entity_candidate_cliques_file, ee_file)
def retrieve_mentions():
line_docs_file_name = 'e:/dc/nyt-world-full/processed/docs-tokenized.txt'
illegal_start_words_file = 'e:/dc/20ng_bydate/stopwords.txt'
output_dir = 'e:/dc/nyt-world-full/processed/mentions/'
dst_doc_entity_candidates_list_file = output_dir + 'doc_mention_candidates.txt'
dst_entity_candidate_clique_file = output_dir + 'mention_candidate_cliques.txt'
dst_doc_entity_indices_file = output_dir + 'doc_mention_candidate_indices.txt'
# init_entity_net(line_docs_file_name, illegal_start_words_file, dst_doc_entity_candidates_list_file,
# dst_entity_candidate_clique_file, dst_doc_entity_indices_file)
lc_word_cnts_file_name = 'e:/dc/el/wiki/wiki_word_cnts_lc.txt'
wc_word_cnts_file_name = 'e:/dc/el/wiki/wiki_word_cnts_with_case.txt'
proper_entity_dict_file = output_dir + 'entity_names.txt'
# gen_entity_name_dict(dst_doc_entity_candidates_list_file, lc_word_cnts_file_name, wc_word_cnts_file_name,
# proper_entity_dict_file)
de_file_bin = 'e:/dc/nyt-world-full/processed/bin/de.bin'
de_file_txt = 'e:/dc/nyt-world-full/processed/bin/de.txt'
dataarange.gen_doc_entity_pairs(proper_entity_dict_file, dst_doc_entity_candidates_list_file, de_file_bin,
de_file_txt)
entity_candidate_cliques_file = dst_entity_candidate_clique_file
ee_file = 'e:/dc/nyt-world-full/processed/bin/ee.bin'
dataarange.gen_entity_entity_pairs(proper_entity_dict_file, entity_candidate_cliques_file, ee_file)
def setup_entity_pairs_file():
doc_list_file = 'e:/dc/20ng_bydate/all_doc_path_list.txt'
docs_ner_file = 'e:/dc/20ng_bydate/docs-for-ner.txt'
# pack_docs_for_ner(doc_list_file, docs_ner_file)
docs_ner_file = 'e:/dc/20ng_bydate/docs-for-ner.txt'
ner_result_file = 'e:/dc/20ng_bydate/ner-result.txt'
cooccur_mentions_file = 'e:/dc/20ng_bydate/cooccur-mentions.txt'
# dataarange.gen_ee_pairs_with_ner_result(docs_ner_file, ner_result_file, cooccur_mentions_file)
# gen entity name dict
entity_name_dict_file = 'e:/dc/20ng_bydate/entity-names-ner.txt'
# dataarange.gen_entity_name_dict(ner_result_file, entity_name_dict_file)
ner_result_file = 'e:/dc/20ng_bydate/ner-result.txt'
doc_all_mentions_file = 'e:/dc/20ng_bydate/doc-mentions-ner.txt'
# dataarange.ner_result_to_tab_sep(ner_result_file, doc_all_mentions_file)
name_dict_file = 'e:/dc/20ng_bydate/entity-names-ner.txt'
doc_entity_file = 'e:/dc/20ng_bydate/bin/de-ner.bin'
dataarange.gen_doc_entity_pairs(name_dict_file, doc_all_mentions_file, doc_entity_file)
entity_candidate_cliques_file = 'e:/dc/20ng_bydate/cooccur-mentions.txt'
ee_file = 'e:/dc/20ng_bydate/bin/ee-ner.bin'
dataarange.gen_entity_entity_pairs(name_dict_file, entity_candidate_cliques_file, ee_file)
cnts_file = 'e:/dc/20ng_bydate/bin/entity-cnts-ner.bin'
dataarange.gen_cnts_file(doc_entity_file, cnts_file)
def __setup_entity_pairs_file():
doc_list_file = 'e:/dc/20ng_bydate/all_doc_path_list.txt'
docs_ner_file = 'e:/dc/20ng_bydate/docs-for-ner.txt'
pack_docs_for_ner(doc_list_file, docs_ner_file)
docs_ner_file = 'e:/dc/20ng_bydate/docs-for-ner.txt'
ner_result_file = 'e:/dc/20ng_bydate/ner-result.txt'
cooccur_mentions_file = 'e:/dc/20ng_bydate/cooccur-mentions.txt'
# dataarange.gen_ee_pairs_with_ner_result(docs_ner_file, ner_result_file, cooccur_mentions_file)
# gen entity name dict
entity_name_dict_file = 'e:/dc/20ng_bydate/entity-names-ner.txt'
# dataarange.gen_entity_name_dict(ner_result_file, entity_name_dict_file)
ner_result_file = 'e:/dc/20ng_bydate/ner-result.txt'
doc_all_mentions_file = 'e:/dc/20ng_bydate/doc-mentions-ner.txt'
# dataarange.ner_result_to_tab_sep(ner_result_file, doc_all_mentions_file)
name_dict_file = 'e:/dc/20ng_bydate/entity-names-ner.txt'
doc_entity_file = 'e:/dc/20ng_bydate/bin/de-ner.bin'
dataarange.gen_doc_entity_pairs(name_dict_file, doc_all_mentions_file,
doc_entity_file)
entity_candidate_cliques_file = 'e:/dc/20ng_bydate/cooccur-mentions.txt'
ee_file = 'e:/dc/20ng_bydate/bin/ee-ner.bin'
dataarange.gen_entity_entity_pairs(name_dict_file,
entity_candidate_cliques_file, ee_file)
cnts_file = 'e:/dc/20ng_bydate/bin/entity-cnts-ner.bin'
dataarange.gen_cnts_file(doc_entity_file, cnts_file)
def __setup_doc_entities_file():
illegal_start_words_file = 'e:/data/emadr/20ng_bydate/stopwords.txt'
emadr_data_dir = 'e:/data/emadr/el/tac/2014/eval/'
line_docs_file = os.path.join(emadr_data_dir, 'docs-tokenized.txt')
doc_entity_candidates_list_file = os.path.join(
emadr_data_dir, 'doc_entity_candidates.txt')
entity_candidate_clique_file = os.path.join(
emadr_data_dir, 'entity_candidate_cliques.txt')
dataarange.init_entity_net(line_docs_file, illegal_start_words_file,
doc_entity_candidates_list_file,
entity_candidate_clique_file)
proper_entity_dict_file = 'e:/data/emadr/el/wiki/entity_names.txt'
# doc_entity_candidates_file = 'e:/data/emadr/el/wiki/doc_entity_candidates.txt'
dst_doc_entity_list_file = os.path.join(emadr_data_dir, 'de.bin')
dataarange.gen_doc_entity_pairs(proper_entity_dict_file,
doc_entity_candidates_list_file,
dst_doc_entity_list_file)