Add Baseline Model Comparison & Improve README

Readme.md

@@ -2,7 +2,7 @@
 
 ## 📌 Project Overview
 QuickFactChecker is a machine learning–based web app that helps detect whether a news article is **real** or **fake**.  
-It uses different models (Naive Bayes, LSTM, etc.) trained on the **LIAR dataset** to evaluate credibility and assist users in identifying potentially misleading information.  
+It uses different models (e.g., Naive Bayes, LSTM) trained on the **LIAR dataset** to evaluate credibility and assist users in identifying potentially misleading information.  
 
 ---
 
@@ -18,7 +18,7 @@ It uses different models (Naive Bayes, LSTM, etc.) trained on the **LIAR dataset
 ## 📂 Project Structure
 ```bash
 dataset/liar
-   ├── README
+   ├── README.md   ##Dataset description
    ├── train.tsv   ##Training data
    ├── test.tsv    ##Testing data
    ├── valid.tsv   ##Validation data
@@ -28,8 +28,16 @@ module/
    ├── fake-news-detection-using-lstm.ipynb
    ├── fake-news-detection-using-nb.ipynb
    ├── liar-data-analysis.ipynb
+
 templates/
    ├── index.html
+
+scripts/
+   └── fake_news_logreg_rf.py     ## Train & evaluate Naive Bayes, Logistic Regression, Random Forest
+results/
+   ├── model_comparison.md        ## Generated baseline comparison table (markdown)
+   └── comparison.png             ## Generated accuracy bar chart
+
 .gitattributes
 app.py
 hero_img.svg
@@ -40,23 +48,42 @@ requirements.txt
 
 ## ⚙️ Installation & Setup
 
-1. Clone the repository:
+1. Clone the repository and navigate into it:
    ```bash
    git clone https://github.com/Deepika14145/QuickFactChecker.git
    cd QuickFactChecker
    ```
 2. Create virtual environment (optional but recommended)
    ```bash
       python -m venv venv
+3. Activate the virtual environment:
+   ```bash
       source venv/bin/activate   # for Linux/Mac
       venv\Scripts\activate      # for Windows
    ```
 
-3. Install the required dependencies:
+4. Install the required dependencies:
    ```bash
    pip install -r requirements.txt
    ```
 
+## 📊 Baseline Model Comparison
+
+We evaluated three models on the LIAR dataset using TF-IDF features. Example results (accuracy & precision):
+example:
+| Model               | Accuracy | Precision |
+|---------------------|----------|-----------|
+| Naive Bayes         | 0.XXXX   | 0.XXXX    |
+| Logistic Regression | 0.XXXX   | 0.XXXX    |
+| Random Forest       | 0.XXXX   | 0.XXXX    |
+
+Logistic Regression achieved the highest accuracy among the tested baselines.
+### 🔧 Run the comparison script
+To reproduce these results, run:
+```bash
+scripts/fake_news_logreg_rf.py
+```
+
 ## Usage
 
 1. Run the following command to start the application:
@@ -67,6 +94,7 @@ requirements.txt
 2. The app will provide predictions on whether a news article is real or fake based on the input.
 
 ## 🛠️ Model Training
+To retrain or experiment with the models, run the provided Jupyter notebooks. Ensure your virtual environment is activated and all dependencies are installed.
 ### Naive Bayes
 Run the notebook:
  ```bash
@@ -90,7 +118,7 @@ Contributions are welcome! Follow these steps:
 1. Fork the repository
 2. Create a new branch (git checkout -b feature-name)
 3. Make your changes
-4. Commit your changes (git commit -m 'Add feature')
+4. Commit your changes (git commit -m 'description of your feature/fix')
 5. Push to the branch  (git push origin feature-name)
 6. Create a Pull Request
 
@@ -100,7 +128,7 @@ Please read [CONTRIBUTING.md](CONTRIBUTING.md) and follow our [Code of Conduct](
 
 For queries, feedback, or guidance regarding this project, you can contact the **mentor** assigned to the issue:  
 
-- 📩 **GitHub**: [Deepika14145](https://github.com/Deepika14145)
+- 📩 **GitHub**: [Deepika14145](https://github.com/Deepika14145)(owner of this repository)
 - 💬 **By commit/PR comments**: Please tag the mentor in your commit or pull request discussion for direct feedback.  
 
 Original Repository: [QuickFactChecker](https://github.com/Deepika14145/QuickFactChecker.git)  
@@ -112,6 +140,6 @@ This project is licensed under the **MIT License** - see the [LICENSE](LICENSE)
 
 --- 
 
-If you like this project, please give it a ⭐ star. Your support means a lot to us!
+If you find this project useful, please give it a ⭐️! Your support is appreciated.!
 
 Feel free to contribute or suggest new features!🙏

results/model_comparison.md

@@ -0,0 +1,7 @@
+# Model Comparison Results
+
+| Model               | Accuracy| Precision |
+|---------------------|---------|-----------|
+| Naive Bayes         | 0.2344  | 0.3558    |
+| Logistic Regression | 0.2402  | 0.2328    |
+| Random Forest       | 0.2310  | 0.2386    |

scripts/fake_news_logreg_rf.py

@@ -0,0 +1,91 @@
+import pandas as pd
+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.metrics import accuracy_score, precision_score
+import matplotlib.pyplot as plt
+import os
+
+# -------------------------
+# Load and preprocess dataset
+# -------------------------
+df = pd.read_csv("QuickFactChecker/dataset/liar/train.tsv", sep="\t", on_bad_lines="warn")
+
+df.columns = [
+    "id", "label", "statement", "subject", "speaker", "job", "state", "party",
+    "barely_true_counts", "false_counts", "half_true_counts", "mostly_true_counts",
+    "pants_on_fire_counts", "context"
+]
+
+X = df["statement"]   # claim text
+y = df["label"]       # truth label (categories)
+
+# Convert text into TF-IDF features
+vectorizer = TfidfVectorizer(max_features=5000, stop_words="english")
+X_vec = vectorizer.fit_transform(X)
+
+# Split dataset
+X_train, X_test, y_train, y_test = train_test_split(
+    X_vec, y, test_size=0.2, random_state=42
+)
+
+# -------------------------
+# Helper function for training
+# -------------------------
+def train_and_evaluate(model, name, results):
+    model.fit(X_train, y_train)
+    y_pred = model.predict(X_test)
+    acc = accuracy_score(y_test, y_pred)
+    prec = precision_score(y_test, y_pred, average="macro", zero_division=0)
+    results[name] = {"accuracy": acc, "precision": prec}
+
+# -------------------------
+# Train models
+# -------------------------
+results = {}
+
+train_and_evaluate(MultinomialNB(), "Naive Bayes", results)
+train_and_evaluate(LogisticRegression(max_iter=1000), "Logistic Regression", results)
+train_and_evaluate(RandomForestClassifier(n_estimators=100, random_state=42), "Random Forest", results)
+
+# -------------------------
+# Print results in table
+# -------------------------
+print("\nModel Performance Comparison:\n")
+print("{:<20} {:<10} {:<10}".format("Model", "Accuracy", "Precision"))
+for model, scores in results.items():
+    print("{:<20} {:.4f}    {:.4f}".format(model, scores["accuracy"], scores["precision"]))
+
+# -------------------------
+# Save results to markdown
+# -------------------------
+os.makedirs("results", exist_ok=True)
+
+with open("results/model_comparison.md", "w") as f:
+    f.write("# Model Comparison Results\n\n")
+    f.write("| Model              | Accuracy | Precision |\n")
+    f.write("|--------------------|----------|-----------|\n")
+    for model, scores in results.items():
+        f.write(f"| {model} | {scores['accuracy']:.4f} | {scores['precision']:.4f} |\n")
+
+# -------------------------
+# Plot comparison
+# -------------------------
+models = list(results.keys())
+accuracies = [results[m]["accuracy"] for m in models]
+
+plt.figure(figsize=(8,5))
+plt.bar(models, accuracies, color=['skyblue', 'lightgreen', 'salmon'])
+plt.ylim(0,0.5)
+plt.xlabel("Models")
+plt.ylabel("Accuracy")
+plt.title("Model Accuracy Comparison")
+
+# Add accuracy labels
+for i, acc in enumerate(accuracies):
+    plt.text(i, acc + 0.01, f"{acc:.2f}", ha='center', fontsize=12)
+
+plt.savefig("results/comparison.png")
+plt.show()