Upload app.py

app.py

@@ -63,9 +63,16 @@ def generate_insights(cluster_characteristics):
         insights.append("High click-through rate: Users are interacting well with ads. Increase ad relevance to boost conversions.")
     if cluster_characteristics['Bounce Rate'] > 0.3:
         insights.append("High bounce rate: Review landing page design and content relevance to improve user retention.")
+    if cluster_characteristics['Frequency of Visits'] > 15:
+        insights.append("Frequent visits: Users are returning often, consider loyalty programs or personalized content to maintain engagement.")
+    if cluster_characteristics['Time Spent per Page'] < 20:
+        insights.append("Low time spent per page: Content may not be engaging or relevant enough. Consider content optimization.")
+    if cluster_characteristics['Conversion Rate'] > 0.15:
+        insights.append("High conversion rate: Effective ad targeting. Explore scaling up ad spend on similar user segments.")
     return " ".join(insights)
 
 def predict_cluster(session_duration, pages_visited, ads_clicked, engagement_score, user_interests, device_type, time_of_day, time_spent_per_page, click_through_rate, conversion_rate, frequency_of_visits, bounce_rate):
+    logging.info("Starting cluster prediction.")
     input_df = pd.DataFrame({
         'Session Duration': [session_duration],
         'Pages Visited': [pages_visited],
@@ -80,7 +87,9 @@ def predict_cluster(session_duration, pages_visited, ads_clicked, engagement_sco
         'Frequency of Visits': [frequency_of_visits],
         'Bounce Rate': [bounce_rate]
     })
+    logging.debug(f"Input DataFrame: {input_df}")
     cluster = pipeline.predict(input_df)[0]
+    logging.info(f"Predicted cluster: {cluster}")
     centroids = pipeline.named_steps['cluster'].cluster_centers_
     cluster_characteristics = centroids[cluster]
 
@@ -99,19 +108,22 @@ def predict_cluster(session_duration, pages_visited, ads_clicked, engagement_sco
     # Generate actionable insights
     insights = generate_insights(cluster_characteristics)
 
+    logging.info("Cluster prediction completed.")
     return f"Predicted Cluster: {cluster}\nCharacteristics: {cluster_characteristics}\nActionable Insights: {insights}"
 
 def ad_performance_analytics():
-    # Calculate average metrics
+    logging.info("Calculating ad performance analytics.")
     avg_ctr = data['Click Through Rate'].mean()
     avg_conversion_rate = data['Conversion Rate'].mean()
     avg_bounce_rate = data['Bounce Rate'].mean()
+    logging.debug(f"Average CTR: {avg_ctr}, Average Conversion Rate: {avg_conversion_rate}, Average Bounce Rate: {avg_bounce_rate}")
     
     # Prepare the analytics report
     report = f"Average Click Through Rate: {avg_ctr:.2%}\n"
     report += f"Average Conversion Rate: {avg_conversion_rate:.2%}\n"
     report += f"Average Bounce Rate: {avg_bounce_rate:.2%}"
     
+    logging.info("Ad performance analytics calculation completed.")
     return report
 
 with gr.Blocks() as demo:
@@ -140,6 +152,7 @@ with gr.Blocks() as demo:
             ],
             outputs=output_textbox
         )
+        logging.info("Gradio predict button configured.")
 
     with gr.Tab("Ad Performance Analytics"):
         gr.Markdown("""
@@ -157,6 +170,8 @@ with gr.Blocks() as demo:
             ad_performance_analytics,
             outputs=analytics_output
         )
+        logging.info("Gradio analytics button configured.")
 
 demo.launch()
+logging.info("Gradio interface launched.")