def test_different_initials(self):
assert_true(who.ratio(self.name, 'E. R. Liebowitz') < 100)
assert_true(who.ratio(self.name, 'E. Liebowitz') < 100)
assert_true(who.ratio(self.name, 'R. V. Liebowitz') < 100)
assert_true(who.ratio(self.name, 'O. E. Liebowitz') < 100)
assert_true(who.ratio(self.name, 'E. R. Liebowitz') <
who.ratio(self.name, 'E. E. Liebowitz'))
assert_true(who.ratio(self.name, 'E. R. Liebowitz') <
who.ratio(self.name, 'R. R. Liebowitz'))
assert_true(who.ratio(self.name, 'E. R. Liebowitz') <
who.ratio(self.name, 'E. Liebowitz'))
def test_nickname(self):
name = 'Robert "Evan" Liebowitz'
assert_equal(who.ratio(name, 'Evan Liebowitz'), 100)
assert_equal(who.ratio('Evan Liebowitz', name), 100)
assert_true(who.ratio(name, 'Wrongbert Lieobwitz') < 100)
assert_true(who.ratio(name, 'Robert Evan') < 100)
assert_true(who.ratio(name, 'Evan Liebowitz',
options={'check_nickname': False}) < 100)
assert_true(who.ratio(name, 'xxxx Liebowitz') <
who.ratio(name, 'xvax Liebowitz'))
assert_equal(who.ratio(name, 'xxxx Liebowitz'),
who.ratio(name, 'xvax Liebowitz', 'strict'))
def string_disambiguation(value, candidates, name=False):
def get_label(x):
return x.split('resource')[1][1:].replace('_', ' ')
if name:
best_dist, best_match = (-1, -float('inf')), None
for entity in candidates:
label = get_label(entity)
name_sim = who.ratio(label, value)
neg_lev_dist = -levenshtein(value.lower(), label.lower())
dist = (name_sim, neg_lev_dist)
if dist[1] <= 0 and dist > best_dist:
best_dist = dist
best_match = entity
else:
distances = []
best_dist, best_match = 999999, None
for entity in candidates:
label = get_label(entity)
dist = levenshtein(value.lower(), label.lower(), best_dist)
if 0 <= dist < best_dist:
best_dist = dist
best_match = entity
return best_match
def test_suffixes(self):
name = 'Robert Liebowitz Jr'
assert_equal(who.ratio(name, 'Robert Liebowitz'), 100)
assert_equal(who.ratio(name, 'Robert Liebowitz Jr'), 100)
assert_equal(who.ratio(name, 'Robert Liebowitz, PhD'), 100)
assert_false(who.ratio(name, 'Robert Liebowitz, Sr'))
assert_false(who.ratio(name, 'Robert Liebowitz, Sr, PhD'))
assert_equal(who.ratio(name, 'Robert Liebowitz, Jr, PhD'), 100)
# Suffix doesn't change a match
assert_equal(who.ratio(name, 'Zachary Liebowitz, Jr'),
who.ratio(name, 'Zachary Liebowitz'))
def test_titles(self):
name = 'Mr. Robert Liebowitz'
assert_equal(who.ratio(name, 'Robert Liebowitz'), 100)
assert_equal(who.ratio(name, 'Sir Robert Liebowitz'), 100)
assert_equal(who.ratio(name, 'Dr. Robert Liebowitz'), 100)
assert_false(who.ratio(name, 'Mrs. Robert Liebowitz'))
# Title doesn't change a match
assert_equal(who.ratio(name, 'Dr. Zachary Liebowitz'),
who.ratio(name, 'Zachary Liebowitz'))
s = p.extract(s1, choices, limit=len(choices))
q = p.extract(s2, choices, limit=len(choices))
# a flag
error = 0
# aggregate count for all the names in list when compared with 1st reference name
d1 = 0
print(s, q)
for i in s:
d1 += i[1]
# if score is somewhere less than 70, use whoswho library
# since if acronym is there for some name then whoswho gives better result
if i[1] < 70:
q1 = (w.ratio(s1, i[0]))
q2 = (w.ratio(s2, i[0]))
print(i[0])
# if score is still <60 then block the user
if q1 < 60 and q2 < 60:
print('blocked')
error = 1
break
else:
# if score >=60
# decrement fuzzywuzzy score and add whoswho score
d1 -= i[1]
d1 += max(q1, q2)
error = 0
# is all the names passed successfully from 1st reference name, repeat same process for second reference name
def test_name_and_initials(self):
assert_equal(who.ratio(self.name, 'R. Evan Liebowitz'), 100)
assert_equal(who.ratio(self.name, 'Robert E. Liebowitz'), 100)
assert_equal(who.ratio(self.name, 'R. E. Liebowitz'), 100)
def test_unicode(self):
name = self.name
assert_equal(who.ratio(name, 'attaché Robert Evan Liebowitz'), 100)
assert_equal(who.ratio(name, 'Rōbért Èvān Lîęböwitz'), 100)
assert_true(who.ratio(name, 'Rōbért Èvān Lęîböwitz') < 100)
def test_string(self):
# Only relevant for python 2.X
assert_equal(who.ratio(self.name, str('Robert Liebowitz')), 100)
def test_equivalent_suffixes(self):
name = 'Robert Liebowitz Jr'
assert_equal(who.ratio(name, 'Robert Liebowitz Jnr'), 100)
assert_false(who.ratio(name, 'Robert Liebowitz Snr'))
def test_short_names(self):
assert_true(who.ratio(self.name, 'Rob Liebowitz'))
assert_true(who.ratio(self.name, 'Bert Liebowitz') < 100)
assert_true(who.ratio(self.name, 'Robbie Liebowitz') < 100)
assert_true(who.ratio(self.name, 'xxxxx Liebowitz') <
who.ratio(self.name, 'Bobby Liebowitz'))
def test_different_number_initials(self):
assert_equal(who.ratio(self.name, 'Robert Liebowitz'), 100)
assert_equal(who.ratio(self.name, 'R. Liebowitz'), 100)
assert_true(who.ratio(self.name, 'Robert E. E. Liebowitz') < 100)
assert_true(who.ratio(self.name, 'R. E. E. Liebowitz') < 100)
assert_equal(who.ratio('R.E.E. Liebowitz', 'R. E. E. Liebowitz'), 100)
def index():
name1 = request.args.get('name1')
name2 = request.args.get('name2')
return str(who.ratio(name1, name2))
from queue import PriorityQueue
import pandas as pd
from whoswho import who
from tqdm import tqdm
from itertools import count
repec_authors = pd.read_csv('csv/authors.csv')
top50_authors = pd.read_csv('csv/top50.csv')
tiebreaker = count()
match_results = []
for index, row in tqdm(top50_authors.iterrows(), total=top50_authors.shape[0]):
name = row['Name']
sorted_matches = PriorityQueue()
for i, r in tqdm(repec_authors.iterrows(), total=repec_authors.shape[0]):
repec_name = r['author_name']
score = who.ratio(name, repec_name, 'strict')
# print(score, r)
repect_author = {'score': score}
repect_author.update(r.to_dict())
sorted_matches.put((-score, next(tiebreaker), repect_author))
match_result = row.to_dict()
for i in range(1, 4):
match = sorted_matches.get()[-1]
for key in match.keys():
match_result[f'match_{i}_{key}'] = match[key]
match_results.append(match_result)
df = pd.DataFrame(match_results)
df.to_csv('csv/top50_matched.csv', index=False)
