import os
import subprocess
# Install required libraries if missing
required_libs = ["streamlit", "pandas", "numpy", "scikit-learn", "nltk"]
for lib in required_libs:
subprocess.run(["pip", "install", lib])
import streamlit as st
import pandas as pd
import numpy as np
import nltk
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import MultinomialNB
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import RandomForestClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score
from nltk.corpus import stopwords
st.image("innomatics-footer-logo.webp")
st.image("fake_logo.jpg")
# Download NLTK stopwords
nltk.download("stopwords")
stop_words = set(stopwords.words("english"))
# Load Datasets
@st.cache_data
def load_data():
df_fake = pd.read_csv("Fake.csv")
df_real = pd.read_csv("True.csv")
# Assign labels
df_fake["label"] = 0 # Fake News
df_real["label"] = 1 # Real News
# Merge datasets
df = pd.concat([df_fake, df_real], ignore_index=True)
df = df.sample(n=10000, random_state=27).reset_index(drop=True) # Shuffle
return df
df = load_data()
# Text Preprocessing Function
def preprocess_text(text):
text = text.lower()
text = " ".join(word for word in text.split() if word not in stop_words)
return text
df["clean_text"] = df["text"].astype(str).apply(preprocess_text)
# TF-IDF Vectorization
vectorizer = TfidfVectorizer(max_features=2000)
X = vectorizer.fit_transform(df["clean_text"])
# Target variable
y = df["label"].values
# Split Data
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# Train Models
models = {
"Logistic Regression": LogisticRegression(),
"Naive Bayes": MultinomialNB(),
"Random Forest": RandomForestClassifier(n_estimators=100, random_state=42),
"Decision Tree": DecisionTreeClassifier(random_state=42),
}
# Streamlit App UI
st.markdown("📰 Fake News Detection App
", unsafe_allow_html=True)
st.markdown("Select a machine learning model and enter a news article to predict if it's Real or Fake.
", unsafe_allow_html=True)
# Model Selection Dropdown
st.markdown("🔍 Choose a Machine Learning Model:
", unsafe_allow_html=True)
selected_model = st.selectbox("", list(models.keys()))
# Train Selected Model
model = models[selected_model]
model.fit(X_train, y_train)
# Predictions
y_pred = model.predict(X_test)
# Classification Report
accuracy = accuracy_score(y_test, y_pred)
precision = precision_score(y_test, y_pred)
recall = recall_score(y_test, y_pred)
f1 = f1_score(y_test, y_pred)
# Display Model Performance
st.markdown("📊 Model Performance
", unsafe_allow_html=True)
st.write(f"Accuracy: {accuracy:.4f}", unsafe_allow_html=True)
st.write(f"Precision: {precision:.4f}", unsafe_allow_html=True)
st.write(f"Recall: {recall:.4f}", unsafe_allow_html=True)
st.write(f"F1 Score: {f1:.4f}", unsafe_allow_html=True)
# User Input
# Styled Text Area Label
st.markdown("📝 Enter News Article:
", unsafe_allow_html=True)
# Text Area for User Input
news_input = st.text_area("", height=200)
# Function to Predict News Authenticity
def predict_news(article, model):
clean_text = preprocess_text(article)
text_features = vectorizer.transform([clean_text]).toarray()
prediction = model.predict(text_features)[0]
return "🟢 Real News" if prediction == 1 else "🔴 Fake News"
if st.button("Check News Authenticity"):
if news_input.strip() == "":
st.warning("⚠ Please enter a news article before clicking the button.")
else:
result = predict_news(news_input, model)
st.markdown("Prediction Result:
", unsafe_allow_html=True)
st.markdown(f"{result}
", unsafe_allow_html=True)