def save_tweets():
connect_to_mongo()
tweets_by_emotion = get_clean_tweets(EMOTION_CLASS_MAP,
NUM_CLEAN_TWEETS_PER_CLASS)
for emotion, tweets in tweets_by_emotion.items():
for tweet in tweets:
print("Fetched Tweet ID - " + tweet.get("id_str"))
try:
tweet_model = Tweet.objects(id_str=tweet.get("id_str")).get()
except DoesNotExist:
tweet_model = Tweet(id_str=tweet.get("id_str"))
tweet_model.text = tweet.get("text")
tweet_model.emotion_label = emotion
tweet_model.created_at = date_parse(tweet.get("created_at"))
# get rid of duplications
try:
tweet_model.save()
except NotUniqueError:
print(
f"Tweet {tweet_model.id_str} alredy exists - will discard......."
)
import json
import pandas as pd
from sklearn.metrics import confusion_matrix, classification_report
from source.database_classes import connect_to_mongo, Tweet, ProcessedTweet
#this python returns the classification report with the statistical analysis of the data fetched and processed from twitter
connect_to_mongo()
initial_df = pd.read_csv("data/tweets_saved.csv")
initial_df = initial_df[["emotion_label", "id_str"]]
initial_df['id_str'] = initial_df['id_str'].astype(str)
results_df = pd.read_csv("data/tweets_processed.csv")
results_df = results_df[["emotion_label", "id_str"]]
results_df['id_str'] = results_df['id_str'].astype(str)
merged = pd.merge(initial_df, results_df, on="id_str")
merged["id_str"] = merged["id_str"].apply(str)
SENTIMENT_TO_EMOTION = {
"Excitement": "excitement",
"Happiness": "happy",
"Fear": "fear",
"Surprise": "surprise",
"Pleasant": "pleasant",
"Anger": "anger",
}