-
Notifications
You must be signed in to change notification settings - Fork 1
/
3_scattertext.py
180 lines (150 loc) · 5.45 KB
/
3_scattertext.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
"""Runs scattertext on the cleaned abstracts"""
import pandas as pd
import scattertext as st
import src.general as gen
def one_vs_rest_authors(author_data) -> None:
"""Run one-vs-rest Scattertext analysis for authors"""
corpus = (
st.CorpusFromPandas(
author_data,
category_col="Full_name",
text_col="Abstract_bigram_ws",
nlp=st.whitespace_nlp,
)
.build()
.compact(st.AssociationCompactor(2000))
)
for author in author_data["Full_name"].unique():
print(f">>>> {author}")
html = st.produce_scattertext_explorer(
corpus,
category=author,
category_name=author,
not_category_name="Other Authors",
minimum_term_frequency=2,
pmi_filter_thresold=4,
transform=st.Scalers.dense_rank,
width_in_pixels=1000,
metadata=author_data["Title"],
include_gradient=True,
left_gradient_term=f"Less like {author}",
right_gradient_term=f"More like {author}",
)
output_file = gen.OUTPUT_DIRECTORY / f"{author}_scattertext.html"
with open(output_file, "wb") as outfile:
outfile.write(html.encode("utf-8"))
def one_vs_rest_journals(journal_data) -> None:
"""Run one-vs-rest Scattertext analysis for journals"""
corpus = (
st.CorpusFromPandas(
journal_data,
category_col="Journal",
text_col="Abstract_bigram_ws",
nlp=st.whitespace_nlp,
)
.build()
.compact(st.AssociationCompactor(2000))
)
for journal in journal_data["Journal"].unique():
print(f">>>> {journal}")
html = st.produce_scattertext_explorer(
corpus,
category=journal,
category_name=journal,
not_category_name="Other Journals",
minimum_term_frequency=20,
pmi_filter_thresold=4,
transform=st.Scalers.dense_rank,
width_in_pixels=1000,
include_gradient=True,
left_gradient_term=f"Less like {journal}",
right_gradient_term=f"More like {journal}",
)
output_file = gen.OUTPUT_DIRECTORY / f"{journal}_scattertext.html"
with open(output_file, "wb") as outfile:
outfile.write(html.encode("utf-8"))
def citation_extremes(journal_data) -> None:
"""Scattertext of words associated with the most and fewest citations"""
# No citations yet for 2024 articles
journal_data = journal_data[journal_data["Pub Date"] < 2024]
def get_most_least_cited(group) -> pd.DataFrame:
threshold = group["Citations"].quantile(0.80)
group["Most_cited"] = group["Citations"].apply(
lambda x: "high" if x >= threshold else "not"
)
threshold = group["Citations"].quantile(0.20)
group["Least_cited"] = group["Citations"].apply(
lambda x: "low" if x <= threshold else "not"
)
return group
journal_data = (
journal_data.groupby("Pub Date")
.apply(get_most_least_cited)
.reset_index(drop=True)
)
corpus = (
st.CorpusFromPandas(
journal_data,
category_col="Most_cited",
text_col="Abstract_bigram_ws",
nlp=st.whitespace_nlp,
)
.build()
.compact(st.AssociationCompactor(2000, use_non_text_features=False))
)
html = st.produce_scattertext_explorer(
corpus,
category="high",
category_name="Most Cited",
not_category_name="Not Most Cited",
minimum_term_frequency=20,
pmi_filter_thresold=4,
transform=st.Scalers.dense_rank,
width_in_pixels=1000,
include_gradient=True,
left_gradient_term="Less likely to be most cited",
right_gradient_term="More likely to be most cited",
)
output_file = gen.OUTPUT_DIRECTORY / "Most_cited_scattertext.html"
with open(output_file, "wb") as outfile:
outfile.write(html.encode("utf-8"))
corpus = (
st.CorpusFromPandas(
journal_data,
category_col="Least_cited",
text_col="Abstract_bigram_ws",
nlp=st.whitespace_nlp,
)
.build()
.compact(st.AssociationCompactor(2000, use_non_text_features=False))
)
html = st.produce_scattertext_explorer(
corpus,
category="low",
category_name="Least Cited",
not_category_name="Not Least Cited",
minimum_term_frequency=20,
pmi_filter_thresold=4,
width_in_pixels=1000,
include_gradient=True,
left_gradient_term="Less likely to be least cited",
right_gradient_term="More likely to be most cited",
)
output_file = gen.OUTPUT_DIRECTORY / "Least_cited_scattertext.html"
with open(output_file, "wb") as outfile:
outfile.write(html.encode("utf-8"))
def main() -> None:
"""Main function for scattertext analysis"""
print("=====Loading data files into memory=====")
author_data, journal_data = gen.load_data_files()
print("=====Running scattertext analyses=====")
print(">> Authors...")
one_vs_rest_authors(author_data)
print(">> Journals...")
one_vs_rest_journals(journal_data)
print(">> Citation extremes...")
citation_extremes(journal_data)
print("*****Processing complete*****")
print(f"Scattertexts saved to {gen.OUTPUT_DIRECTORY}")
if __name__ == "__main__":
main()