Twitter_Sentiment_Analysis.py

import streamlit as st
import twint
import pandas as pd
import numpy as np
from wordcloud import WordCloud
import matplotlib.pyplot as plt
import preprocess
import VADER
import TextBlobModel
import LogisticRegression
import SVM
import NaiveBayes



def Visualize(model):
	df = pd.read_csv(model + '_Sentiments.csv', error_bad_lines=False, engine='python', encoding = 'utf8')
	graph, ax = plt.subplots()
	d = df['Polarity'].value_counts()
	d = d.sort_index(ascending=False)
	ax = d.plot.bar(x='Polarity', rot=0)
	for p in ax.patches:
		ax.annotate(np.round(p.get_height(),decimals=2),
             			    (p.get_x()+p.get_width()/2., p.get_height()), 
                                     ha='center', va='center', xytext=(0, 10), textcoords='offset points')
	st.pyplot(graph)
	if(model == 'TextBlob'):
		fig1, ax = plt.subplots(figsize=(8,6)) 
		for i in range(0, df.shape[0]):
			ax.scatter(df["Polarity_Score"][i], df["Subjectivity"][i], color='Blue')
		ax.set_title('Sentiment Analysis') 
		ax.set_xlabel('Polarity')
		ax.set_ylabel('Subjectivity')
		st.pyplot(fig1) 

	# word cloud visualization
	plt.style.use('fivethirtyeight')
	df = df.dropna(subset=['tweet'])
	allWords = ' '.join([twts for twts in df['tweet']])
	wordCloud = WordCloud(width=1000, height=500, random_state=21, max_font_size=110).generate(allWords)
	plt.imshow(wordCloud, interpolation="bilinear")
	plt.axis('off')
	fig = plt.figure()
	st.image(wordCloud.to_array())

	st.dataframe(df)
	
st.sidebar.title("About")
st.sidebar.info(
    "Web App that Extracts latest Tweets based on Input **Search Term** & Performs Sentiment Analysis on the extracted tweets\n\n"
    "This app is created and maintained by [Ajinkya Chavan](https://github.com/AjinkyaChavan9)\n\n"
    "Check the [Source Code] (https://github.com/AjinkyaChavan9/Twitter-Sentiment-Analysis)"
)

st.sidebar.title("Contribute")
st.sidebar.info(
	"You are very **Welcome** to contribute your awesome comments, questions or suggestions as [issues](https://github.com/AjinkyaChavan9/Twitter-Sentiment-Analysis/issues) "
	"or [pull requests](https://github.com/AjinkyaChavan9/Twitter-Sentiment-Analysis/pulls) to the [source code](https://github.com/AjinkyaChavan9/Twitter-Sentiment-Analysis)"
)

Title_html = """
    <style>
        .title h1{
          user-select: none;
          font-size: 43px;
          color: white;
          background: repeating-linear-gradient(-45deg, red 0%, yellow 7.14%, rgb(0,255,0) 14.28%, rgb(0,255,255) 21.4%, cyan 28.56%, blue 35.7%, magenta 42.84%, red 50%);
          background-size: 600vw 600vw;
          -webkit-text-fill-color: transparent;
          -webkit-background-clip: text;
          animation: slide 10s linear infinite forwards;
        }
        @keyframes slide {
          0%{
            background-position-x: 0%;
          }
          100%{
            background-position-x: 600vw;
          }
        }
    </style> 
    
    <div class="title">
        <h1>Twitter Sentiment Analysis</h1>
    </div>
    """
st.markdown(Title_html, unsafe_allow_html=True) #Title rendering

## User Input
search_term = st.text_input("Enter Search Term to Analyze Tweets:")
no_of_tweets = st.number_input("Enter Number of Tweets:",  value=500, step=100)

# Configure Twint
c = twint.Config()
c.Search = search_term
c.Lang = "en"
c.Limit = no_of_tweets
c.Store_csv = True
c.Custom["tweet"] = ["username", "tweet"]
c.Output = search_term + '_tweets_data.csv'

if st.button("Generate Dataset"):
	#Run Twint
	twint.run.Search(c)
	df = pd.read_csv(search_term + '_tweets_data.csv', error_bad_lines=False, engine='python', encoding = 'utf8')
	st.success('Tweets Dataset Generated Successfully!')
	preprocess.preprocess_function(search_term)


selected_model = st.selectbox("Select Model for Sentiment Analysis: ", ['VADER (Valence Aware Dictionary for Sentiment Reasoning) Model', 'TextBlob', 'Logistic Regression', 'SVM', 'Naive Bayes'])

if st.button('Perform Sentiment Analysis'):
	if(selected_model == 'VADER (Valence Aware Dictionary for Sentiment Reasoning) Model'):
		VADER.predict_sentiment_VADER(search_term)
		Visualize('Vader')
	elif(selected_model == 'TextBlob'):
		TextBlobModel.predict_sentiment_TextBlob(search_term)
		Visualize('TextBlob')
	elif(selected_model == 'Logistic Regression'):
		LogisticRegression.predict_sentiment_LogReg(search_term)
		Visualize('Logistic_Regression')
	elif(selected_model == 'SVM'):
		SVM.predict_sentiment_SVM(search_term)
		Visualize('SVM')
	elif(selected_model == 'Naive Bayes'):
		NaiveBayes.predict_sentiment_NB(search_term)
		Visualize('NB')