def test_get_citation_data(self):
'''Test getting citation data'''
from models import get_citation_data
data = get_citation_data(testset)
# The most important thing here is to test that it is a list
# of MetricsModel instances
self.assertEqual(isinstance(data, list), True)
self.assertTrue(False not in
[x.__class__.__name__ == 'MetricsModel' for x in data])
def test_get_citation_data(self):
'''Test getting citation data'''
from models import get_citation_data
data = get_citation_data(testset)
# The most important thing here is to test that it is a list
# of MetricsModel instances
self.assertEqual(isinstance(data, list), True)
self.assertTrue(
False not in [x.__class__.__name__ == 'MetricsModel' for
x in data])
def get_citation_stats(identifiers, bibcodes, bibcodes_ref):
data = selfcits = citdata = None
# citation stats for all publications
cs = {}
# citation stats for refereed publications
csr = {}
# Get the data to compute the citation statistics
# First get data with just the numbers
data = get_citation_data(identifiers)
Nzero = len(bibcodes) - len(data)
Nzero_ref = len(bibcodes_ref) - \
len([p.citation_num for p in data if p.refereed])
citnums = [p.citation_num for p in data] + [0] * Nzero
ref_citnums = [p.refereed_citation_num for p in data] + [0] * Nzero
citnums_ref = [p.citation_num
for p in data if p.refereed] + [0] * Nzero_ref
ref_citnums_ref = [p.refereed_citation_num
for p in data if p.refereed] + [0] * Nzero_ref
# Next, get more detailed citation information
# (with data to be used later on)
# citdata : data structure with citation data for reuse later on
# selfcits : data structure with self-citations
# Nself : number of self-citations
# Nself_ref : number of self-citations for refereed publications
# Nciting : number of citing papers
# Nciting_ref: number of citing papers for refereed publications
citdata, selfcits, Nself, Nself_ref, Nciting, Nciting_ref = \
get_selfcitations(identifiers, bibcodes)
# The number of unique citing papers and the number of self-citations
cs['number of citing papers'] = Nciting
csr['number of citing papers'] = Nciting_ref
cs['number of self-citations'] = Nself
csr['number of self-citations'] = Nself_ref
# The citation stats
# Total number of citations
cs['total number of citations'] = np.sum([p.citation_num for p in data]
or [0])
csr['total number of citations'] = np.sum(
[p.citation_num for p in data if p.refereed] or [0])
# Average number of citations
cs['average number of citations'] = np.mean(citnums or [0])
csr['average number of citations'] = np.mean(citnums_ref or [0])
# Median number of citations
cs['median number of citations'] = np.median(citnums or [0])
csr['median number of citations'] = np.median(citnums_ref or [0])
# Normalized number of citations
cs['normalized number of citations'] = np.sum(
[float(p.citation_num) / float(p.author_num) for p in data] or [0])
csr['normalized number of citations'] = np.sum([
float(p.citation_num) / float(p.author_num) for p in data if p.refereed
] or [0])
# The refereed citations stats
##
cs['total number of refereed citations'] = np.sum(
[p.refereed_citation_num for p in data] or [0])
csr['total number of refereed citations'] = np.sum(
[p.refereed_citation_num for p in data if p.refereed] or [0])
cs['average number of refereed citations'] = np.mean(ref_citnums or [0])
csr['average number of refereed citations'] = np.mean(ref_citnums_ref
or [0])
cs['median number of refereed citations'] = np.median(ref_citnums or [0])
csr['median number of refereed citations'] = np.median(ref_citnums_ref
or [0])
cs['normalized number of refereed citations'] = np.sum(
[float(p.refereed_citation_num) / float(p.author_num) for p in data]
or [0])
csr['normalized number of refereed citations'] = np.sum([
float(p.refereed_citation_num) / float(p.author_num)
for p in data if p.refereed
] or [0])
# Send the results back
return cs, csr, data, selfcits, citdata
def get_citation_stats(identifiers, bibcodes, bibcodes_ref):
data = selfcits = citdata = None
# citation stats for all publications
cs = {}
# citation stats for refereed publications
csr = {}
# Get the data to compute the citation statistics
# First get data with just the numbers
data = get_citation_data(identifiers)
Nzero = len(bibcodes) - len(data)
Nzero_ref = len(bibcodes_ref) - \
len([p.citation_num for p in data if p.refereed])
citnums = [p.citation_num for p in data] + [0] * Nzero
ref_citnums = [p.refereed_citation_num for p in data] + [0] * Nzero
citnums_ref = [
p.citation_num for p in data if p.refereed] + [0] * Nzero_ref
ref_citnums_ref = [
p.refereed_citation_num for p in data if p.refereed] + [0] * Nzero_ref
# Next, get more detailed citation information
# (with data to be used later on)
# citdata : data structure with citation data for reuse later on
# selfcits : data structure with self-citations
# Nself : number of self-citations
# Nself_ref : number of self-citations for refereed publications
# Nciting : number of citing papers
# Nciting_ref: number of citing papers for refereed publications
citdata, selfcits, Nself, Nself_ref, Nciting, Nciting_ref = \
get_selfcitations(identifiers, bibcodes)
# The number of unique citing papers and the number of self-citations
cs['number of citing papers'] = Nciting
csr['number of citing papers'] = Nciting_ref
cs['number of self-citations'] = Nself
csr['number of self-citations'] = Nself_ref
# The citation stats
# Total number of citations
cs['total number of citations'] = np.sum(
[p.citation_num for p in data] or [0])
csr['total number of citations'] = np.sum(
[p.citation_num for p in data if p.refereed] or [0])
# Average number of citations
cs['average number of citations'] = np.mean(citnums or [0])
csr['average number of citations'] = np.mean(citnums_ref or [0])
# Median number of citations
cs['median number of citations'] = np.median(citnums or [0])
csr['median number of citations'] = np.median(citnums_ref or [0])
# Normalized number of citations
cs['normalized number of citations'] = np.sum(
[float(p.citation_num) / float(p.author_num) for p in data] or [0])
csr['normalized number of citations'] = np.sum(
[float(p.citation_num) / float(p.author_num) for p in data if
p.refereed] or [0])
# The refereed citations stats
##
cs['total number of refereed citations'] = np.sum(
[p.refereed_citation_num for p in data]or [0])
csr['total number of refereed citations'] = np.sum(
[p.refereed_citation_num for p in data if p.refereed] or [0])
cs['average number of refereed citations'] = np.mean(ref_citnums or [0])
csr['average number of refereed citations'] = np.mean(
ref_citnums_ref or [0])
cs['median number of refereed citations'] = np.median(ref_citnums or [0])
csr['median number of refereed citations'] = np.median(
ref_citnums_ref or [0])
cs['normalized number of refereed citations'] = np.sum(
[float(p.refereed_citation_num) / float(p.author_num) for
p in data] or [0])
csr['normalized number of refereed citations'] = np.sum(
[float(p.refereed_citation_num) / float(p.author_num) for
p in data if p.refereed] or [0])
# Send the results back
return cs, csr, data, selfcits, citdata
