def most_referenced(sample, amount):
with open('data/02-refined/%s-rows.json' % sample, 'rU') as f:
rows = json.load(f)
for row in rows:
# pprint(row.keys())
for author in row['article']['authors']:
if not authors.has_key(author):
authors[author] = {'articles': 0, 'referenced': 0}
authors[author]['articles'] += 1
for reference in row['article']['referencelist']:
for author in reference['authors']:
if not authors.has_key(author):
authors[author] = {'articles': 0, 'referenced': 0}
authors[author]['referenced'] += 1
pprint(author)
with open('data/03-stats/%s-top-references.csv' % sample, 'wb') as f:
writer = CSVUnicodeWriter(f)
for row in rows:
for reference in row['article']['referencelist']:
for author in reference['authors']:
for top_author, stats in sorted(
authors.items(),
key=lambda x: x[1]['referenced'],
reverse=True)[:amount]:
if author == top_author:
writer.writerow([
','.join(reference['authors']),
reference['year'], reference['original']
])
def references_by_authors_db(sample, authors):
f = open('data/03-stats/%s-top-references.csv' % sample, 'wb')
writer = CSVUnicodeWriter(f)
for author, stats in authors:
print author
for reference in db.references.find({
'articleset': sample,
'authors': author
}):
# pprint(reference)
writer.writerow([
','.join(reference['authors']), reference['year'],
reference['original']
])
f.close()
