Anusheh/moderator flow

.gitignore

@@ -1,3 +1,5 @@
 *.log
 DiscordBot/tokens.json
 TABot/tokens.json
+*.csv
+*.joblib

DiscordBot/classifier/misinfo_classifier.py

@@ -0,0 +1,161 @@
+import pandas as pd
+import numpy as np
+from sklearn.model_selection import train_test_split, GridSearchCV, cross_val_score
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.linear_model import LogisticRegression
+from sklearn.metrics import classification_report, accuracy_score
+from sklearn.pipeline import Pipeline
+import nltk
+from nltk.corpus import stopwords
+from nltk.tokenize import word_tokenize
+import re
+import joblib
+import os
+
+# Download required NLTK data
+nltk.download('punkt')
+nltk.download('stopwords')
+
+def preprocess_text(text):
+    """Preprocess text by removing special characters, converting to lowercase, and removing stopwords."""
+    if isinstance(text, str):
+        # Convert to lowercase
+        text = text.lower()
+        # Remove special characters and digits
+        text = re.sub(r'[^a-zA-Z\s]', '', text)
+        # Tokenize
+        tokens = word_tokenize(text)
+        # Remove stopwords
+        stop_words = set(stopwords.words('english'))
+        tokens = [token for token in tokens if token not in stop_words]
+        return ' '.join(tokens)
+    return ''
+
+def load_and_preprocess_data():
+    """Load and preprocess the dataset."""
+    # Load the datasets
+    fake_df = pd.read_csv('DataSet_Misinfo_FAKE.csv')
+    true_df = pd.read_csv('DataSet_Misinfo_TRUE.csv')
+
+    # Rename columns for consistency
+    fake_df.columns = ['index', 'text']
+    true_df.columns = ['index', 'text']
+
+    # Add labels
+    fake_df['label'] = 1  # 1 for fake/misinformation
+    true_df['label'] = 0  # 0 for true
+
+    # Combine datasets
+    df = pd.concat([fake_df, true_df], ignore_index=True)
+
+    # Preprocess text
+    df['processed_text'] = df['text'].apply(preprocess_text)
+
+    return df
+
+def train_classifier(save_model=True):
+    """Train and evaluate the misinformation classifier with cross-validation and hyperparameter tuning."""
+    # Load and preprocess data
+    df = load_and_preprocess_data()
+
+    # Split data into training and testing sets
+    X_train, X_test, y_train, y_test = train_test_split(
+        df['processed_text'],
+        df['label'],
+        test_size=0.2,
+        random_state=42
+    )
+
+    # Create a pipeline with TF-IDF vectorizer and classifier
+    pipeline = Pipeline([
+        ('tfidf', TfidfVectorizer()),
+        ('clf', LogisticRegression())
+    ])
+
+    # Define hyperparameter grid
+    param_grid = {
+        'tfidf__max_features': [3000, 5000, 7000],
+        'tfidf__ngram_range': [(1, 1), (1, 2)],
+        'clf__C': [0.1, 1.0, 10.0],
+        'clf__max_iter': [1000]
+    }
+
+    # Perform grid search with cross-validation
+    grid_search = GridSearchCV(
+        pipeline,
+        param_grid,
+        cv=5,  # 5-fold cross-validation
+        scoring='accuracy',
+        n_jobs=-1  # Use all available CPU cores
+    )
+
+    print("Performing grid search with cross-validation...")
+    grid_search.fit(X_train, y_train)
+
+    # Get best parameters and score
+    print("\nBest parameters:", grid_search.best_params_)
+    print("Best cross-validation score:", grid_search.best_score_)
+
+    # Evaluate on test set
+    best_model = grid_search.best_estimator_
+    y_pred = best_model.predict(X_test)
+
+    print("\nTest Set Classification Report:")
+    print(classification_report(y_test, y_pred))
+    print("\nTest Set Accuracy:", accuracy_score(y_test, y_pred))
+
+    # Perform additional cross-validation on the best model
+    cv_scores = cross_val_score(best_model, X_train, y_train, cv=5)
+    print("\nCross-validation scores:", cv_scores)
+    print("Mean CV score:", cv_scores.mean())
+    print("CV score std:", cv_scores.std())
+
+    if save_model:
+        # Create models directory if it doesn't exist
+        os.makedirs('models', exist_ok=True)
+
+        # Save the best model and vectorizer
+        model_path = 'models/misinfo_classifier.joblib'
+        joblib.dump(best_model, model_path)
+        print(f"\nModel saved to {model_path}")
+
+    return best_model
+
+def load_model(model_path='classifier/models/misinfo_classifier.joblib'):
+    """Load a trained model from file."""
+    if not os.path.exists(model_path):
+        raise FileNotFoundError(f"Model file not found at {model_path}")
+    return joblib.load(model_path)
+
+def predict_misinformation(text, model):
+    """Predict if a given text is misinformation or not."""
+    # Make prediction using the loaded model
+    prediction = model.predict([text])[0]
+    probability = model.predict_proba([text])[0]
+
+    return {
+        'is_misinformation': bool(prediction),
+        'confidence': float(probability[prediction]),
+        'true_probability': float(probability[0]),
+        'fake_probability': float(probability[1])
+    }
+
+if __name__ == "__main__":
+    # # Train the classifier
+    # print("Training classifier...")
+    # model = train_classifier(save_model=True)
+
+    # # Example usage with loaded model
+    # test_text = "This is an example text to test the classifier."
+    # result = predict_misinformation(test_text, model)
+
+    # print("\nExample prediction:")
+    # print(f"Text: {test_text}")
+    # print(f"Is misinformation: {result['is_misinformation']}")
+    # print(f"Confidence: {result['confidence']:.2f}")
+    # print(f"True probability: {result['true_probability']:.2f}")
+    # print(f"Fake probability: {result['fake_probability']:.2f}") 
+
+    model = load_model()
+    prediction = predict_misinformation("Russia did not interfere in the 2016 presidential election", model)
+    print(prediction)

DiscordBot/classifier/requirements.txt

@@ -0,0 +1,4 @@
+pandas>=1.3.0
+numpy>=1.21.0
+scikit-learn>=0.24.2
+nltk>=3.6.0