adds CleanTweets functions, creates Graphs

analyze.py

@@ -5,6 +5,7 @@ import pandas as pd
 from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline
 from datasets import load_dataset
 from transformers.pipelines.pt_utils import KeyDataset
+from funs.CleanTweets import remove_URL, remove_emoji, remove_html, remove_punct
 
 
 #%%
@@ -82,33 +83,6 @@ model = AutoModelForSequenceClassification.from_pretrained("bvrau/covid-twitter-
 tokenizer = AutoTokenizer.from_pretrained("bvrau/covid-twitter-bert-v2-struth")
 
 # Source https://www.kaggle.com/code/daotan/tweet-analysis-with-transformers-bert
-def remove_URL(text):
-    url = re.compile(r'https?://\S+|www\.\S+')
-    return url.sub(r'', text)
-
-
-def remove_emoji(text):
-    emoji_pattern = re.compile(
-        '['
-        u'\U0001F600-\U0001F64F'  # emoticons
-        u'\U0001F300-\U0001F5FF'  # symbols & pictographs
-        u'\U0001F680-\U0001F6FF'  # transport & map symbols
-        u'\U0001F1E0-\U0001F1FF'  # flags (iOS)
-        u'\U00002702-\U000027B0'
-        u'\U000024C2-\U0001F251'
-        ']+',
-        flags=re.UNICODE)
-    return emoji_pattern.sub(r'', text)
-
-
-def remove_html(text):
-    html = re.compile(r'<.*?>|&([a-z0-9]+|#[0-9]{1,6}|#x[0-9a-f]{1,6});')
-    return re.sub(html, '', text)
-
-
-def remove_punct(text):
-    table = str.maketrans('', '', string.punctuation)
-    return text.translate(table)
 
 dfPreTest['cleanContent'] = dfPreTest['rawContent'].apply(remove_URL)
 dfPreTest['cleanContent'] = dfPreTest['cleanContent'].apply(remove_emoji)

cleanTweets.py

@@ -8,7 +8,7 @@ Created on Mon Jun 26 20:36:43 2023
 
 import pandas as pd
 # import pyreadstat
-# import numpy as np
+import numpy as np
 from funs.ClearDupes import deDupe
 
 
@@ -32,11 +32,13 @@ senCSV = "ALL-SENATORS-TWEETS.csv"
 senDataset = "senators-raw.csv"
 
 # Name of new datafile generated
-senCSVc = "Tweets-Cleaned"
+senCSVc = "SenatorsTweets-Final"
+senCSVcCov = "SenatorsTweets-OnlyCov"
 
 # don't change this one
 senCSVPath = wd + ud + senCSV
-senCSVcPath = wd + ud + senCSV + ".csv"
+senCSVcPath = wd + ud + senCSVc + ".csv"
+senCSVcCovPath = wd + ud + senCSVcCov + ".csv"
 senSAVcPath = wd + ud + senCSV + ".sav"
 senDTAcPath = wd + ud + senCSV + ".dta"
 senDatasetPath = wd + di + senDataset
@@ -90,21 +92,17 @@ with open(f"{di}keywords.txt", "w") as file:
 
 #%%
 # overwrite keyword column
-df['contains_keyword'] = ''
-df['contains_keyword'] = (
-    df['rawContent'].str.findall('|'.join(keywords)).str.join(',').replace('', 'none')
+df['keywords'] = np.nan
+df['keywords'] = (
+    df['rawContent'].str.findall('|'.join(keywords)).str.join(',').replace('', np.nan)
 )
-mask = (df['contains_keyword'] != 'none') # select all values in contains_keyword == 'none'
-df.loc[mask,'keywords'] = df['contains_keyword'] # set keywords = contains_keyword under the condition of mask
 #%%
-# create bool contains_keyword
-df['contains_keyword'] = ~pd.isnull(df['keywords']) # create boolean column
-#%%
-# recode contains keyword to bool
-mask = (df['contains_keyword'] != 'none')
-df.loc[mask,'contains_keyword'] = True
-df.loc[~mask,'contains_keyword'] = False # ~ negates mask, selecting all values that do not contain keywords
+# create boolean contains_keyword column
+df['contains_keyword'] = True
+mask = (df['keywords'].isna()) # select all values in contains_keyword == 'none'
+df.loc[mask,'contains_keyword'] = False # set keywords = contains_keyword under the condition of mask
 
+#%%
 pd.Series(df["user.id"]).is_unique
 
 #%%
@@ -157,10 +155,14 @@ dfAll = df.merge(dfSenAll, how='left',on='user.username')
 unique_usernames = dfAll.loc[dfAll['name'].isnull(), 'user.username'].unique()
 print(unique_usernames)
 # senatorisakson was dropped, is ok
+#%%
+# create covidtweets csv
+dfCov = dfAll[dfAll['contains_keyword']==True]
 
 #%%
 # Export to csv, sav and dta
 dfAll.to_csv(senCSVcPath, encoding='utf-8')
+dfCov.to_csv(senCSVcCovPath, encoding='utf-8')
 # pyreadstat.write_sav(df, senSAVcPath) # commented out because file generated is 11 gb 
 # =============================================================================
 # dfAll.rename(columns=lambda x: x.replace('.', '_'), inplace=True)

createGraphs.py

@@ -0,0 +1,144 @@
+#%%
+#!/usr/bin/env python3
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+from wordcloud import WordCloud
+from funs.CleanTweets import remove_URL, remove_emoji, remove_html, remove_punct
+import string
+#%%
+
+# -*- coding: utf-8 -*-
+"""
+Created on Mon Jun 26 20:36:43 2023
+
+@author: michael
+"""
+
+import pandas as pd
+# import pyreadstat
+# import numpy as np
+
+###################
+# Setup directories
+# WD Michael
+wd = "/home/michael/Documents/PS/Data/collectTweets/"
+# WD Server
+# wd = '/home/yunohost.multimedia/polsoc/Politics & Society/TweetCollection/'
+
+# datafile input directory
+di = "data/IN/"
+
+# Tweet-datafile output directory
+ud = "data/OUT/"
+
+# Name of file that all senator data will be written to
+senCSV = "SenatorsTweets-OnlyCov.csv" # SenatorsTweets-Final.csv SenatorsTweets-OnlyCov.csv
+
+# Name of file that all senator data will be written to
+senDataset = "senators-raw.csv"
+
+# Name of new datafile generated
+senCSVc = "SenatorsTweets-Final.csv"
+senCSVcCov = "SenatorsTweets-OnlyCov.csv"
+
+# Outfiles
+wcAllTweetsF = "graphs/Wordcloud-All.png"
+wcCovTweetsF = "graphs/Wordcloud-Cov.png"
+TwCovTimeline = "graphs/Timeline.png"
+
+# don't change this one
+senCSVcPath = wd + ud + senCSVc
+senCSVcCovPath = wd + ud + senCSVcCov
+wcAllTweetsFPath = wd + ud + wcAllTweetsF
+wcCovTweetsFPath = wd + ud + wcCovTweetsF
+TwCovTimelinePath = wd + ud + TwCovTimeline
+
+#%%
+df = pd.read_csv(senCSVcPath, dtype=(object))
+dfCov = pd.read_csv(senCSVcCovPath, dtype=(object))
+#%%
+df['cleanContent'] = df['rawContent'].apply(remove_URL)
+df['cleanContent'] = df['cleanContent'].apply(remove_emoji)
+df['cleanContent'] = df['cleanContent'].apply(remove_html)
+df['cleanContent'] = df['cleanContent'].apply(remove_punct)
+
+# create string with all cleaned tweets as text
+str_alltweets = df['cleanContent'].astype(str).str.cat(sep=' ').casefold()
+#%%
+dfCov['cleanContent'] = dfCov['rawContent'].apply(remove_URL)
+dfCov['cleanContent'] = dfCov['cleanContent'].apply(remove_emoji)
+dfCov['cleanContent'] = dfCov['cleanContent'].apply(remove_html)
+dfCov['cleanContent'] = dfCov['cleanContent'].apply(remove_punct)
+
+# create string with all cleaned tweets as text
+str_covtweets = dfCov['cleanContent'].astype(str).str.cat(sep=' ').casefold()
+#%%
+# replace single U and S characters
+str_covtweets = str_covtweets.replace(' u ', ' ') 
+str_covtweets = str_covtweets.replace(' s ', ' ') 
+str_alltweets = str_alltweets.replace(' u ', ' ') 
+str_alltweets = str_alltweets.replace(' s ', ' ') 
+
+
+# %%
+# create wordcloud alltweets
+wcA = WordCloud(background_color="white", width=1000, height=1000, repeat=True)
+wcA.generate(str_alltweets)
+
+#%%
+# draw
+plt.figure( figsize=(20,20))
+plt.axis("off")
+plt.imshow(wcA, interpolation="bilinear")
+fig1 = plt.gcf()
+plt.show()
+fig1.savefig(wcAllTweetsFPath)  
+
+# %%
+# create wordcloud covtweets
+wcC = WordCloud(background_color="white", width=1000, height=1000, repeat=True)
+wcC.generate(str_covtweets)
+#%%
+# draw
+plt.figure( figsize=(20,20))
+plt.axis("off")
+plt.imshow(wcC, interpolation="bilinear")
+fig2 = plt.gcf()
+plt.show()
+fig2.savefig(wcCovTweetsFPath)  
+# %%
+# with open('test.txt', 'w') as f:
+#    f.write(str_covtweets)
+# %%
+dfT = pd.DataFrame()
+dfT['date'] = df['date'].copy()
+dfT['count'] = 1
+
+dfCovT = pd.DataFrame()
+dfCovT['date'] = dfCov['date'].copy()
+dfCovT['count'] = 1
+#%%
+dfT['date'] = pd.to_datetime(dfT['date']).dt.strftime('%Y-%m-%d')
+dfCovT['date'] = pd.to_datetime(dfCovT['date']).dt.strftime('%Y-%m-%d')
+
+#%%
+dfT = dfT.groupby('date').count().reset_index()
+dfCovT = dfCovT.groupby('date').count().reset_index()
+
+#%%
+import matplotlib.dates as mdates
+# n of tweets overall
+my_dpi=300
+plt.figure(figsize=(1000/my_dpi, 1500/my_dpi), dpi=my_dpi)
+plt.style.use('seaborn-darkgrid')
+fig, ax = plt.subplots(figsize=(8, 6))
+ax.plot(dfCovT['date'], dfCovT['count'], marker='', color='tab:blue', linewidth=1, alpha=0.4)
+ax.plot(dfT['date'], dfT['count'], marker='', color='tab:blue', linewidth=1, alpha=1)
+ax.xaxis.set_major_locator(mdates.MonthLocator(interval=3))
+ax.xaxis.set_minor_locator(mdates.MonthLocator())
+fig.autofmt_xdate()
+fig.savefig(TwCovTimelinePath)  
+
+
+# %%

funs/CleanTweets.py

@@ -0,0 +1,39 @@
+import re
+import string
+
+def remove_URL(text):
+    url = re.compile(r'https?://\S+|www\.\S+')
+    return url.sub(r'', text)
+
+
+def remove_emoji(text):
+    emoji_pattern = re.compile(
+        '['
+        u'\U0001F600-\U0001F64F'  # emoticons
+        u'\U0001F300-\U0001F5FF'  # symbols & pictographs
+        u'\U0001F680-\U0001F6FF'  # transport & map symbols
+        u'\U0001F1E0-\U0001F1FF'  # flags (iOS)
+        u'\U00002702-\U000027B0'
+        u'\U000024C2-\U0001F251'
+        ']+',
+        flags=re.UNICODE)
+    return emoji_pattern.sub(r'', text)
+
+
+def remove_html(text):
+    html = re.compile(r'<.*?>|&([a-z0-9]+|#[0-9]{1,6}|#x[0-9a-f]{1,6});')
+    return re.sub(html, '', text)
+
+
+def remove_punct(text):
+    table = str.maketrans('', '', string.punctuation)
+    return text.translate(table)
+
+def clean_all(text):
+    if not isinstance(text, str):
+        text = str(text) # Convert non-string values to string
+    text = remove_URL(text)
+    text = remove_emoji(text)
+    text = remove_html(text)
+    text = remove_punct(text)
+    return text