def writeToEventsFile(self, event_clust):
print "Writing Events to a file"
words_list = []
for index, row in event_clust.data.iterrows():
words_list += Cosine_Sim.tokenize_only(row['tweet_text'])
word_counts = Counter(words_list)
most_common = word_counts.most_common(10)
text_file = open("events/Events.txt", "a")
text_file.write("Cluster Id =" + str(event_clust.id) + " ,")
for word, count in most_common:
text_file.write("{0} : {1} ,".format(word, count))
text_file.write("\n")
text_file.close()
# write cluster to csv file
clust_file = 'events/cluster_' + str(event_clust.id) + '.csv'
if os.path.exists(clust_file):
os.remove(clust_file)
try:
event_clust.data.to_csv(clust_file, index=False, encoding='utf-8')
except:
print " Error writing the Event File"
# print "working on tweet ", index
# words_list = Cosine_Sim.tokenize_only(row['tweet_text'])
# words_list_str = " ".join(words_list)
# text_file.write(words_list_str)
# text_file.close()
# cnt = cnt + 1
cnt = 1
cluster_ids = []
files = glob.glob("\clusters_AvgSimilariy\*.csv")
for f in files:
print "working on file , ", f
data = pd.read_csv(f, encoding='utf-8')
cnt_str = f + ","
for index, row in data.iterrows():
text_file = open(r"\tweets_avg\tweet" + str(cnt) + ".txt", "w")
print "working on tweet ", index
words_list = Cosine_Sim.tokenize_only(row['tweet_text'])
words_list_str = " ".join(words_list)
text_file.write(words_list_str)
text_file.close()
cnt_str = cnt_str + str(cnt) + ","
cnt = cnt + 1
cluster_ids.append(cnt_str)
text_file = open(r"\tweets_avg\tweet_clusters.txt", "w")
for i in cluster_ids:
text_file.write(i)
text_file.write("\n")
text_file.close()
