Initial commit: Twitter sentiment analysis with gradio

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "git.ignoreLimitWarning": true
+}

src/.gradio/certificate.pem

@@ -0,0 +1,31 @@
+-----BEGIN CERTIFICATE-----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=
+-----END CERTIFICATE-----

src/.gradio/flagged/dataset1.csv

@@ -0,0 +1,3 @@
+tweet,output,timestamp
+I am very happy,Positive,2025-09-05 19:31:51.696127
+I am very happy,Positive,2025-09-05 19:31:53.644682

src/app.py

@@ -0,0 +1,46 @@
+import gradio as gr
+
+from model import predict_sentiment
+
+GITHUB_LINK = "https://github.com/<your-username>/<your-repo>"
+COFFEE_LINK = "https://www.buymeacoffee.com/samiali"  # <-- replace with your BuyMeACoffee link
+
+
+with gr.Blocks(theme=gr.themes.Citrus()) as app:
+    title="Twitter Sentiment Analysis",
+    gr.Markdown(
+        """
+        # 🌟 Twitter Sentiment Analysis  
+        Enter a tweet below and find out if it's **Positive** or **Negative**.  
+        _Model: Naive Bayes trained on NLTK Twitter samples_
+        """
+    ),
+    with gr.Row():
+        with gr.Column():
+            text = gr.Textbox(
+                placeholder="Type your tweet here...",
+                lines=3,
+                label="Your Tweet"
+            ),
+            btn = gr.Button("🔍 Analyze Sentiment", variant="primary")
+        with gr.Column():
+            output = gr.Label(label="Prediction")
+
+    gr.Markdown(
+        f"""
+        🔗 **Source Code on [GitHub]({GITHUB_LINK})**  
+
+        ☕ If you like this project, consider [buying me a coffee]({COFFEE_LINK})  
+
+        <a href="{COFFEE_LINK}" target="_blank">
+          <img src="https://cdn.buymeacoffee.com/buttons/default-orange.png" 
+               alt="Buy Me A Coffee" height="41" width="174">
+        </a>
+        """
+    )
+
+    gr.Markdown("💾 All predictions are stored for analysis.")
+    btn.click(predict_sentiment, inputs=text, outputs=output)
+
+if __name__ == '__main__':
+    app.launch(share=True)

src/model.py

@@ -0,0 +1,153 @@
+import numpy as np
+import nltk
+import re
+import string
+
+from nltk.corpus import twitter_samples
+from nltk.stem import PorterStemmer
+from nltk.corpus import stopwords
+from nltk.tokenize import TweetTokenizer
+
+nltk.download('twitter_samples')
+nltk.download('stopwords')
+
+positive_tweets = twitter_samples.strings('positive_tweets.json')
+negative_tweets = twitter_samples.strings('negative_tweets.json')
+
+test_pos = positive_tweets[4000:]
+train_pos = positive_tweets[:4000]
+test_neg = negative_tweets[4000:]
+train_neg = negative_tweets[:4000]
+
+train_x = train_pos + train_neg
+test_x = test_pos + test_neg
+
+print(f"Number of positive tweets: {len(positive_tweets)}")
+print(f"Number of negative tweets: {len(negative_tweets)}")
+
+train_y = np.append(np.ones(len(train_pos)), np.zeros(len(train_neg)))
+test_y = np.append(np.ones(len(test_pos)), np.zeros(len(test_neg)))
+
+print("train_y.shape = " + str(train_y.shape))
+print("test_y.shape = " + str(test_y.shape))
+
+
+def process_tweet(tweet):
+  stemmer = PorterStemmer()
+  stopwords_english = stopwords.words('english')
+  tweet = re.sub(r'\$\w*', '', tweet)
+  tweet = re.sub(r'^RT[\s]+', '', tweet)
+  tweet = re.sub(r'https?:\/\/.*[\r\n]*', '', tweet)
+  tweet = re.sub(r'#', '', tweet)
+  tokenizer = TweetTokenizer(preserve_case=False, strip_handles=True,
+                               reduce_len=True)
+  tweet_tokens = tokenizer.tokenize(tweet)
+
+  tweets_clean = []
+  for word in tweet_tokens:
+      if (word not in stopwords_english and
+          word not in string.punctuation):
+          stem_word = stemmer.stem(word)
+          tweets_clean.append(stem_word)
+
+  return tweets_clean
+
+
+print("Before tweet processing: ", positive_tweets[0])
+print("After tweet processing: ", process_tweet(positive_tweets[0]))
+
+def build_freqs(tweets, ys):
+  freq_dict = {}
+  for tweet, y in zip(tweets, ys):
+    tweet = process_tweet(tweet)
+    for word in tweet:
+      if (word, y) in freq_dict:
+        freq_dict[(word, y)] += 1
+      else:
+        freq_dict[(word, y)] = 1
+  return freq_dict
+
+# create frequency dictionary
+freqs = build_freqs(train_x, train_y)
+
+# check the output
+print("type(freqs) = " + str(type(freqs)))
+print("len(freqs) = " + str(len(freqs.keys())))
+
+def train_naive_bayes(freq, train_x, train_y):
+  vocab = set([pair[0] for pair in freq.keys()])
+  V = len(vocab)
+  loglikelihood = {}
+  logprior = 0
+
+  N_pos, N_neg = 0, 0
+  V_pos, V_neg = 0, 0
+
+  for pair in freq.keys():
+    if pair[1] > 0.0:
+      N_pos += freq[pair]
+      V_pos += 1
+    else:
+      N_neg += freq[pair]
+      V_pos += 1
+
+  D = len(train_y)
+
+  D_pos = len(list(filter(lambda x: x > 0, train_y)))
+  D_neg = len(list(filter(lambda x: x <= 0, train_y)))
+
+  logprior = np.log(D_pos) - np.log(D_neg)
+
+  for word in vocab:
+    freq_pos = freq.get((word, 1.0), 0)
+    freq_neg = freq.get((word, 0.0), 0)
+
+    temp_pos_prob = (freq_pos + 1) / (N_pos + V)
+    temp_neg_prob = (freq_neg + 1) / (N_neg + V)
+
+    loglikelihood[word] = np.log(temp_pos_prob / temp_neg_prob)
+
+  return logprior, loglikelihood
+
+
+logprior, loglikelihood = train_naive_bayes(freqs, train_x, train_y)
+
+
+def predict(tweet, logprior, loglikelihood):
+  word_l = process_tweet(tweet)
+  p = 0
+  p += logprior
+  for word in word_l:
+    if word in loglikelihood:
+      p += loglikelihood[word]
+  return p
+
+my_tweet = 'She smiled.'
+p = predict(my_tweet, logprior, loglikelihood)
+print('The expected output is', p)
+
+def evaluate(test_x, test_y, logprior, loglikelihood):
+  accuracy = 0
+  y_hats = []
+  for tweet in test_x:
+    y_hat = predict(tweet, logprior, loglikelihood)
+    if y_hat > 0:
+      y_hat_i = 1
+    else:
+      y_hat_i = 0
+    y_hats.append(y_hat_i)
+  accuracy = np.absolute(np.mean(np.equal(test_y, y_hats)))
+  return accuracy
+
+print("Naive Bayes accuracy = %0.4f" %
+      (evaluate(test_x, test_y, logprior, loglikelihood)))
+
+def predict_sentiment(tweet):
+  p = predict(tweet, logprior, loglikelihood)
+
+  if p > 1:
+    return "Positive"
+  elif p >= 0 and p <= 1:
+    return "Neutral"
+  else:
+    return "Negative"