mkallipo
/
affRo
1
1
Fork 0
Code Issues Pull Requests Packages Projects Releases 1 Wiki Activity

updates

This commit is contained in:
mkallipo 2024-10-28 11:13:55 +01:00
parent 8c6f6a5a9a
commit 415b45e3ca
10 changed files with 10 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
dictionaries/dix_acad.json
View File

File diff suppressed because one or more lines are too long

1
dictionaries/dix_acad_ext.json
View File

File diff suppressed because one or more lines are too long

2
dictionaries/dix_categ.json
View File

@ -1 +1 @@
{"research": "Univ/Inst", "recherche": "Univ/Inst", "uniwersytet": "Univ/Inst", "investigacions": "Univ/Inst", "institu": "Univ/Inst", "istitut": "Univ/Inst", "univ": "Univ/Inst", "col": "Univ/Inst", "center": "Univ/Inst", "polytechnic": "Univ/Inst", "politecnico": "Univ/Inst", "polutekhneio": "Univ/Inst", "centre": "Univ/Inst", "kentro": "Univ/Inst", "cnrs": "Univ/Inst", "faculty": "Univ/Inst", "school": "Univ/Inst", "academ": "Univ/Inst", "akadem": "Univ/Inst", "hochschule": "Univ/Inst", "ecole": "Univ/Inst", "tech": "Univ/Inst", "observ": "Univ/Inst", "escuela": "Univ/Inst", "escola": "Univ/Inst", "discovery programe": "Univ/Inst", "ku leuven": "Univ/Inst", "ucla": "Univ/Inst", "eth zurich": "Univ/Inst", "athena": "Univ/Inst", "openaire": "Univ/Inst", "erasmus": "Univ/Inst", "ist austria": "Univ/Inst", "lmu munich": "Univ/Inst", "cancer trials ireland": "Univ/Inst", "food safety authority": "Univ/Inst", "ucd": "Univ/Inst", "tcd": "Univ/Inst", "apc microbiome": "Univ/Inst", "nasa": "Univ/Inst", "ucl": "Univ/Inst", "zentrum": "Univ/Inst", "ncsr demokritos": "Univ/Inst", "panepist": "Univ/Inst", "nui galway": "Univ/Inst", "nui maynooth": "Univ/Inst", "tu wien": "Univ/Inst", "tu dublin": "Univ/Inst", "lab": "Laboratory", "science": "Laboratory", "cientific": "Laboratory", "hospital": "Hospital", "clinic": "Hospital", "hopital": "Hospital", "klinik": "Hospital", "oncol": "Hospital", "medical": "Hospital", "health": "Hospital", "medicin": "Hospital", "nosokomei": "Hospital", "gmbh": "Company", "company": "Company", "industr": "Company", "etaireia": "Company", "corporation": "Company", "inc": "Company", "museum": "Museum", "library": "Museum", "foundation": "Foundation", "asociation": "Foundation", "organization": "Foundation", "society": "Foundation", "group": "Foundation", "royal": "Foundation", "ofice": "Foundation", "trust": "Foundation", "district": "Government", "federation": "Government", "government": "Government", "municipal": "Government", "county": "Government", "council": "Government", "agency": "Government", "unknown": "Unknown", "google": "Specific", "yahoo": "Specific", "ebay": "Specific", "microsoft": "Specific", "teagasc": "Specific", "ibm research": "Specific", "alergan": "Specific", "analog devices": "Specific", "medtronic": "Specific", "xilinx": "Specific", "pfizer": "Specific", "glaxosmithkline": "Specific", "astrazenecaboehringer ingelheim": "Specific", "demokritos": "Specific", "siemens": "Specific", "forth": "Specific"}
{"research": "Univ/Inst", "recherche": "Univ/Inst", "uniwersytet": "Univ/Inst", "investigacions": "Univ/Inst", "institu": "Univ/Inst", "istitut": "Univ/Inst", "univ": "Univ/Inst", "col": "Univ/Inst", "center": "Univ/Inst", "polytechnic": "Univ/Inst", "politecnico": "Univ/Inst", "polutekhneio": "Univ/Inst", "centre": "Univ/Inst", "kentro": "Univ/Inst", "cnrs": "Univ/Inst", "faculty": "Univ/Inst", "school": "Univ/Inst", "academ": "Univ/Inst", "akadem": "Univ/Inst", "hochschule": "Univ/Inst", "ecole": "Univ/Inst", "tech": "Univ/Inst", "observ": "Univ/Inst", "escuela": "Univ/Inst", "escola": "Univ/Inst", "discovery programe": "Univ/Inst", "ku leuven": "Univ/Inst", "ucla": "Univ/Inst", "eth zurich": "Univ/Inst", "athena": "Univ/Inst", "openaire": "Univ/Inst", "erasmus": "Univ/Inst", "ist austria": "Univ/Inst", "lmu munich": "Univ/Inst", "cancer trials ireland": "Univ/Inst", "food safety authority": "Univ/Inst", "ucd": "Univ/Inst", "tcd": "Univ/Inst", "apc microbiome": "Univ/Inst", "nasa": "Univ/Inst", "ucl": "Univ/Inst", "zentrum": "Univ/Inst", "ncsr demokritos": "Univ/Inst", "panepist": "Univ/Inst", "nui galway": "Univ/Inst", "nui maynooth": "Univ/Inst", "tu wien": "Univ/Inst", "tu dublin": "Univ/Inst", "lab": "Laboratory", "science": "Laboratory", "cientific": "Laboratory", "hospital": "Hospital", "clinic": "Hospital", "hopital": "Hospital", "klinik": "Hospital", "oncol": "Hospital", "medical": "Hospital", "health": "Hospital", "medicin": "Hospital", "nosokomei": "Hospital", "krankenhaus": "Hospital", "gmbh": "Company", "company": "Company", "industr": "Company", "etaireia": "Company", "corporation": "Company", "inc": "Company", "museum": "Museum", "library": "Museum", "foundation": "Foundation", "asociation": "Foundation", "organization": "Foundation", "society": "Foundation", "group": "Foundation", "royal": "Foundation", "ofice": "Foundation", "trust": "Foundation", "district": "Government", "federation": "Government", "government": "Government", "municipal": "Government", "county": "Government", "council": "Government", "agency": "Government", "unknown": "Unknown", "google": "Specific", "yahoo": "Specific", "ebay": "Specific", "microsoft": "Specific", "teagasc": "Specific", "ibm research": "Specific", "alergan": "Specific", "analog devices": "Specific", "medtronic": "Specific", "xilinx": "Specific", "pfizer": "Specific", "glaxosmithkline": "Specific", "astrazenecaboehringer ingelheim": "Specific", "demokritos": "Specific", "siemens": "Specific", "forth": "Specific", "lily": "Specific", "boeing": "Specific"}

2
dictionaries/dix_city.json
View File

File diff suppressed because one or more lines are too long

1
dictionaries/dix_city_ext.json
View File

File diff suppressed because one or more lines are too long

2
dictionaries/dix_country.json
View File

File diff suppressed because one or more lines are too long

1
dictionaries/dix_country_ext.json
View File

File diff suppressed because one or more lines are too long

2
dictionaries/dix_mult.json
View File

File diff suppressed because one or more lines are too long

1
dictionaries/dix_mult_ext.json
View File

File diff suppressed because one or more lines are too long

8
functions_cluster.py
View File

@ -210,11 +210,11 @@ replacements = {'saint' : 'st',
'universiti':'university',
'universitiy':'university',
'universty' :'university',
'techniche' : 'technological',
'univ col': 'university colege',
'univ. col.': 'university colege',
'univ. coll.': 'university colege',
'col.':'colege',
'army' : 'military',
'hipokration' : 'hipocration',
'belfield, dublin': 'dublin',
'balsbridge, dublin': 'dublin', #ballsbridge
@ -262,10 +262,12 @@ def substrings_dict(string):
modified_value = re.sub(r'universi\w*', 'universi', value, flags=re.IGNORECASE)
modified_value = re.sub(r'institu\w*', 'institu', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'centre*', 'center', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'centre\b', 'center', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'\bsaint\b', 'st', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'\btrinity col\b', 'trinity colege', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'\btechnische\b', 'technological', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'\bteknologi\b', 'technology', modified_value, flags=re.IGNORECASE)
modified_value = re.sub(r'\bpolitehnica\b', 'polytechnic', modified_value, flags=re.IGNORECASE)
@ -293,7 +295,7 @@ def clean_string(input_string):
input_string = input_string.replace(" – ", placeholder)
# Unescape HTML entities and convert to lowercase
input_string = replace_comma_spaces(replace_double_consonants(unidecode(remove_parentheses(html.unescape(input_string.replace("'", ""))))).strip())
input_string = replace_comma_spaces(replace_double_consonants(unidecode(remove_parentheses(html.unescape(input_string.replace(" ́e","e").replace("'", ""))))).strip())
# Replace `–` with space (do not replace hyphen `-`)