app.py

import streamlit as st
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import silhouette_score
import seaborn as sns

st.set_page_config(page_title="Mall Customer Segmentation", layout="wide")

st.title("🛍️ Mall Customer Segmentation using K-Means")

# ----------------------------------------
# Load dataset
# ----------------------------------------
@st.cache_data
def load_data():
    return pd.read_csv("Mall_Customers.csv")

df = load_data()
st.write("### Dataset Preview", df.head())

# ----------------------------------------
# Feature selection
# ----------------------------------------
selected_columns = ['Age', 'Annual Income (k$)', 'Spending Score (1-100)']

st.write("### Pairplot (Age, Income, Spending Score)")
fig1 = sns.pairplot(df[selected_columns])
st.pyplot(fig1)

# ----------------------------------------
# Prepare data for clustering
# ----------------------------------------
X = df[['Age', 'Spending Score (1-100)']]
scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

# ----------------------------------------
# Elbow Method
# ----------------------------------------
st.write("### Elbow Method to Find Optimal k")
inertia = []
K = range(1, 11)

for k in K:
    kmeans = KMeans(n_clusters=k, random_state=42)
    kmeans.fit(X_scaled)
    inertia.append(kmeans.inertia_)

fig2, ax2 = plt.subplots()
ax2.plot(K, inertia, 'bo-')
ax2.set_xlabel('Number of clusters (k)')
ax2.set_ylabel('Inertia')
ax2.set_title('Elbow Method For Optimal k')
st.pyplot(fig2)

# ----------------------------------------
# Choose k interactively
# ----------------------------------------
st.sidebar.header("🔢 Select number of clusters")
k = st.sidebar.slider("Choose k (clusters)", 2, 10, 5)

kmeans = KMeans(n_clusters=k, random_state=42)
df['Cluster'] = kmeans.fit_predict(X_scaled)

# ----------------------------------------
# Cluster visualization
# ----------------------------------------
st.write(f"### Cluster Visualization (k={k})")
fig3, ax3 = plt.subplots(figsize=(8,6))
scatter = ax3.scatter(X_scaled[:, 0], X_scaled[:, 1], c=df['Cluster'], cmap='viridis')
ax3.set_xlabel('Age (scaled)')
ax3.set_ylabel('Spending Score (scaled)')
ax3.set_title('Customer Segments (K-Means Clustering)')
plt.colorbar(scatter, ax=ax3, label='Cluster')
st.pyplot(fig3)

# ----------------------------------------
# Silhouette score
# ----------------------------------------
score = silhouette_score(X_scaled, df['Cluster'])
st.success(f"Silhouette Score for k={k}: **{score:.3f}**")

st.write("### Clustered Data Sample")
st.dataframe(df.head())