Add personas

app.py

@@ -4,6 +4,7 @@ import page_attitudes
 import page_demographics
 import page_shopping
 import page_investing
+import page_personas
 import page_tests
 from urllib.parse import quote, unquote
 from datasets import load_dataset
@@ -39,6 +40,8 @@ if st.sidebar.button("Shopping"):
     st.session_state['page'] = 'Shopping'
 if st.sidebar.button("Investing"):
     st.session_state['page'] = 'Investing'
+if st.sidebar.button("Personas"):
+    st.session_state['page'] = 'Personas'
 if st.sidebar.button("Tests"):
     st.session_state['page'] = 'Tests'
 
@@ -58,6 +61,8 @@ elif st.session_state['page'] == 'Shopping':
     page_shopping.show(df)
 elif st.session_state['page'] == 'Investing':
     page_investing.show(df)
+elif st.session_state['page'] == 'Personas':
+    page_personas.show(df)
 elif st.session_state['page'] == 'Tests':
     page_tests.show(df)
 

page_personas.py

@@ -0,0 +1,47 @@
+from sklearn.cluster import KMeans
+from scipy.stats import chisquare
+import streamlit as st
+import pandas as pd
+
+
+@st.cache_data
+def show(df):
+    st.title("Clustering Students to Build Personas")
+    st.write("Clustering based on 36 fields of Likert data")
+    perform_kmeans_clustering(df)
+
+
+def perform_kmeans_clustering(df):
+    # Read the fields from the uploaded Python file
+    likert_flat_fields = [
+        # your list of questions here...
+    ]
+
+    # Select only the relevant columns for clustering
+    df_likert_real_data = df[likert_flat_fields]
+
+    # Drop rows with missing values for a more accurate clustering
+    df_likert_real_data = df_likert_real_data.dropna()
+
+    # Perform k-means clustering to group students into 3 clusters
+    kmeans_real_data = KMeans(n_clusters=3, n_init=10,
+                              random_state=42).fit(df_likert_real_data)
+
+    # Add the cluster labels to the DataFrame
+    df_likert_real_data['Cluster'] = kmeans_real_data.labels_
+
+    # Calculate the mean score for each question in each cluster
+    cluster_means_real_data = df_likert_real_data.groupby(
+        'Cluster').mean().reset_index()
+
+    # Count the number of students in each cluster
+    cluster_counts = df_likert_real_data['Cluster'].value_counts(
+    ).sort_index().reset_index()
+    cluster_counts.columns = ['Cluster', 'Number of Students']
+
+    # Display the tables in Streamlit
+    st.write("Number of Students in Each Cluster:")
+    st.table(cluster_counts)
+
+    st.write("Mean Scores for Each Question in Each Cluster:")
+    st.table(cluster_means_real_data)