def export_words_to_csv(de_words, en_words, exportFile = 'result.csv'):
table = []
for k in range(len(de_words)):
table.append([de_words[k], en_words[k]])
csv_tools.write_table_csv(exportFile, table)
'Link'
]
list_citations = [list_header]
refTable = csv_tools.read_csv_table(listReferences)
# Note: header of refTable is ['ID', 'Number of references', 'Number of references outside selected database', 'Number of self-cited references', 'List ID of references', 'Title', 'Authors', 'Source title', 'Publisher', 'Year', 'Link']
number_of_citations = [0] * len(refTable)
id_of_citing_papers = [
[] for _ in range(len(refTable))
] # initialize array of empty lists, each row contains list of papers that cite this one
# id_of_citing_papers = [[]]*len(refTable) # theoretically this command also creates an array of empty lists, however due to Python management the append/extend for each item affects all others
for row in refTable[1:]:
paper_id = str(row[0])
#print('Processing paper: ', paper_id)
paper_ref = row[4].split(',')
if not paper_ref == [
''
]: # there are some references in the list of papers
for ref in paper_ref:
number_of_citations[int(ref)] += 1
id_of_citing_papers[int(ref)].append(paper_id)
for k in range(1, len(refTable)):
list_citations.append(
[k, number_of_citations[k], ','.join(id_of_citing_papers[k])] +
refTable[k][5:])
csv_tools.write_table_csv(exportFile, list_citations)
#print(list_citations)
print('Analysis finished. Result is written to file ' + exportFile)
sourceTitle_index, abbr_sourceTitle_index,
publisher_index, conference_index
]
for row in table[1:]:
if not row[sourceTitle_index] in processed_source:
authors_split = row[author_index].split(',')
list_authors = [
author.strip() for author in authors_split
]
flag_author_exist = False
for author in list_authors:
if author in filtered_authors:
flag_author_exist = True
if flag_author_exist:
list_sources.append(
[row[k] for k in access_index])
processed_source.append(row[sourceTitle_index])
# This command should be executed after all the papers were collected
del list_sources[0] # remove header line before sorting
list_sources.sort(
key=lambda x: x[0],
reverse=False) # sort by title, note that ID column is not added yet
list_sources.insert(0, list_header)
for k in range(1, len(list_sources)):
list_sources[k].insert(0, k) # add ID
list_sources[k].append('') # add blank column for Rating
csv_tools.write_table_csv(exportFile, list_sources)
#print(list_sources)
print('Analysis finished. Result is written to file ' + exportFile)
paper_title = ref_titles[kk]
paper_authors = ref_authors[kk]
if not paper_title in paperTable_transpose[
1]:
no_ref_ext += 1
else: # reference is in the paper table
ref_id_in_paperTable = paperTable_transpose[
1].index(paper_title)
list_ref_IDs.append(
str(ref_id_in_paperTable))
self_cite_flag = False
for single_author in paper_authors:
if single_author in paperTable_transpose[
2][ref_id_in_paperTable]:
self_cite_flag = True
if self_cite_flag == True:
no_ref_self += 1
list_ref_IDs_string = ','.join(list_ref_IDs)
ref_detail_row = [
id_in_paperTable, no_ref, no_ref_ext,
no_ref_self, list_ref_IDs_string
] + paperTable[id_in_paperTable][1:]
list_references.append(ref_detail_row)
# This command should be executed after all the references were collected
del list_references[0] # remove header line before sorting
list_references.sort(key=lambda x: x[0], reverse=False) # sort by ID
list_references.insert(0, list_header)
csv_tools.write_table_csv(exportFile, list_references)
#print(list_references)
no_refs_to_net[filtered_authors.index(person)] += 1
if not paper_id in temp_list_citations_from_net[filtered_authors.index(ref_author)]:
temp_list_citations_from_net[filtered_authors.index(ref_author)].append(paper_id)
no_citations_from_net[filtered_authors.index(ref_author)] += 1
if not (paper_id, ref_id) in temp_list_ref_pairs_within_net[filtered_authors.index(person)]:
temp_list_ref_pairs_within_net[filtered_authors.index(person)].append((paper_id, ref_id))
no_times_references_to_net[filtered_authors.index(person)] += 1
if not (paper_id, ref_id) in temp_list_cite_pairs_within_net[filtered_authors.index(ref_author)]:
temp_list_cite_pairs_within_net[filtered_authors.index(ref_author)].append((paper_id, ref_id))
no_times_citations_from_net[filtered_authors.index(ref_author)] += 1
matrix_references[filtered_authors.index(person)][filtered_authors.index(ref_author)] += 1
#matrix_citations[filtered_authors.index(ref_author)][filtered_authors.index(person)] += 1 # just the transpose of matrix_references
for k in range(len(filtered_authors)):
list_stats.append([k+1, filtered_authors[k], no_papers[k], no_papers_citing_net[k], no_papers_cited_by_net[k], no_refs_to_net[k], no_citations_from_net[k], no_total_refs[k], no_times_references_to_net[k], no_times_citations_from_net[k]] + matrix_cite_with_rating[k] + matrix_references[k])
# Extend the header
for rating in rating_list:
list_stats[0].append('Citation rating: ' + rating)
for author in filtered_authors:
list_stats[0].append(author + ' (cited by author in row)')
csv_tools.write_table_csv(exportFile, list_stats)
#print(list_stats)
print('Analysis finished. Result is written to file ' + exportFile)
conference_index = headerTable.index('Conference name')
link_index = headerTable.index('DOI')
year_index = headerTable.index('Year')
access_index = [
title_index, author_index, sourceTitle_index,
publisher_index, abbr_sourceTitle_index,
conference_index, year_index, link_index
]
for row in table[1:]:
if not row[title_index] in [
item[0] for item in list_papers
]:
list_papers.append([row[k] for k in access_index])
# This command should be executed after all the papers were collected
del list_papers[0] # remove header line before sorting
list_papers.sort(
key=lambda x: x[0],
reverse=False) # sort by title, note that ID column is not added yet
list_papers.insert(0, list_header)
for k in range(1, len(list_papers)):
list_papers[k].insert(0, k) # add ID
if list_papers[k][3] in sourceTable_transpose[1]:
# paper inherits rating from source
list_papers[k].append(
sourceTable_transpose[5][sourceTable_transpose[1].index(
list_papers[k][3])])
csv_tools.write_table_csv(exportFile, list_papers)
#print(list_papers)
print('Analysis finished. Result is written to file ' + exportFile)
for paper_affi in affi_data[1:]:
list_authors = paper_affi.split(';')
for author in list_authors:
if not author.strip(
) in list_author_affi: # compare both name and affiliation
list_author_affi.append(author.strip())
# This command should be executed after all the author affiliations were collected
list_author_affi.sort()
list_author_combined, list_affi_combined = group_affi_by_author(
list_author_affi)
authorTable = export_table_combined(list_author_combined,
list_affi_combined)
list_header = ['Author', 'Number of affiliations', 'Affiliations']
authorTable.insert(0, list_header)
csv_tools.write_table_csv(exportFile, authorTable)
## Test with one file
#if __name__ == "__main__":
#csvFile = 'TimonRabczuk_citation_by_authors.csv'
#exportFile = 'list_all_authors.csv'
#list_author_affi = []
## For each CSV file
#table = csv_tools.read_csv_table(csvFile)
#headerTable = table[0]
#if 'Authors with affiliations' in headerTable:
#affi_index = headerTable.index('Authors with affiliations')
#affi_data = [row[affi_index] for row in table]
##print('List of authors and affiliations: \n')
##print(affi_data[1922])
##test = parse_author_affi(affi_data[1922])
else:
exportFile = 'all_database.csv'
#merge_scopus_data(dataFolder, exportFile)
merged_table = []
processed_papers = []
for dirpath, dirs, files in os.walk(dataFolder):
for filename in files:
if filename[-4:].upper() == '.CSV':
csvFile = os.path.join(dirpath, filename)
# For each CSV file
print('Processing file ' + csvFile)
table = csv_tools.read_csv_table(csvFile)
headerTable = table[0]
if 'Title' in headerTable:
title_index = headerTable.index('Title')
for row in table[1:]:
if not row[title_index] in processed_papers:
processed_papers.append(row[title_index])
merged_table.append(row)
# This command should be executed after all the papers were collected
merged_table.sort(
key=lambda x: x[title_index], reverse=False
) # sort by title, assume all data files have the same header
merged_table.insert(0, headerTable)
csv_tools.write_table_csv(exportFile, merged_table)
#print(list_papers)
print('Merging finished. Result is written to file ' + exportFile)