add intrroduction

pages/ALE.py

@@ -52,10 +52,10 @@ def explain_example():
     
     st.pyplot(fig)
 
-    fig, ax = plt.subplots(figsize=(10, 5))
+    fig1, ax1 = plt.subplots(figsize=(10, 5))
     st.write("2D Second-Order ALE Plot")
     ale_plot(global_model, X_test, X_train.columns[:2], bins=10)
-    st.pyplot(fig)
+    st.pyplot(fig1)
 
 def main():
     global global_model

pages/PermutationFeatureImportance.py

@@ -63,8 +63,8 @@ def main():
     if global_model is None:
         train_model()
     
-    st.title("ALE (Accumulated Local Effects)")
-    st.write(prompt_params.PERMUTTATION_INTRODUCTION)
+    st.title("Permutation Feature Importance")
+    st.write(prompt_params.PERMUTATION_INTRODUCTION)
     # Explain the selected sample
     if st.button("Explain Sample"):
         explain_example()

src/prompt_config.py

@@ -1,7 +1,24 @@
 APP_INTRODUCTION = """
-This application provides explainability for machine learning models using LIME and SHAP.
-It allows users to explore how different features influence model predictions by selecting
-specific samples and visualizing their explanations interactively.
+# **Explanatory AI for Income Prediction - App Overview**
+
+This application provides an **Explainable AI (XAI)** framework to analyze a machine learning model trained on the **UCI Adult Income Dataset** ([link](https://archive.ics.uci.edu/dataset/2/adult)). The model predicts whether an individual earns more than **$50,000 per year** based on key demographic and employment-related features, including:
+
+- **Age**
+- **Education Level**
+- **Marital Status**
+- **Sex (Gender)**
+- **Occupation, Work Hours, Country**, etc.
+
+To ensure transparency and interpretability, the app utilizes multiple **XAI techniques** to explain model predictions:
+
+### **Explainability Methods**
+1. **ALE (Accumulated Local Effects)** – Measures feature influence while considering feature dependencies.
+2. **Anchors** – Provides high-precision rule-based explanations for individual predictions.
+3. **ICE & PDP (Individual Conditional Expectation & Partial Dependence Plots)** – Visualizes how a feature affects the model prediction globally and individually.
+4. **LIME (Local Interpretable Model-agnostic Explanations)** – Generates local approximations to explain specific predictions.
+5. **Permutation Feature Importance** – Assesses the importance of features by measuring the change in prediction error after shuffling feature values.
+6. **SHAP (SHapley Additive exPlanations)** – Computes fair feature attributions based on cooperative game theory.
+
 """
 LIME_INTRODUCTION = """
 LIME (Local Interpretable Model-agnostic Explanations) is a technique used to interpret the predictions of black-box machine learning models.