Create app.py

app.py

@@ -0,0 +1,96 @@
+import gradio as gr
+import pandas as pd
+from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
+import requests
+
+# Load results
+df = pd.read_csv('final_results.csv')
+analyzer = SentimentIntensityAnalyzer()
+
+# Your n8n webhook URL
+N8N_WEBHOOK_URL = "https://andreeaalbulescu.app.n8n.cloud/webhook-test/book-price-optimizer"
+
+def analyze_book(book_title):
+    match = df[df['title'].str.contains(book_title, case=False, na=False)]
+    
+    if len(match) > 0:
+        row = match.iloc[0]
+        sentiment = row['avg_sentiment']
+        recommendation = row['recommendation']
+        current_price = row['price']
+        recommended_price = row['recommended_price']
+        rating = row['rating']
+        
+        if sentiment > 0.2:
+            sentiment_label = "Positive 😊"
+        elif sentiment < -0.1:
+            sentiment_label = "Negative 😞"
+        else:
+            sentiment_label = "Neutral 😐"
+        
+        result = f"""
+📚 BOOK ANALYSIS RESULTS
+========================
+Title: {row['title']}
+Current Price: £{current_price}
+Rating: {'⭐' * int(rating)}
+
+📊 SENTIMENT ANALYSIS
+Sentiment: {sentiment_label}
+Score: {round(sentiment, 3)}
+
+💰 PRICING RECOMMENDATION
+Action: {recommendation}
+Recommended Price: £{recommended_price}
+        """
+        
+        # Send to n8n webhook
+        try:
+            requests.post(N8N_WEBHOOK_URL, json={
+                "book": book_title,
+                "sentiment": round(float(sentiment), 3),
+                "recommendation": recommendation,
+                "current_price": float(current_price),
+                "recommended_price": float(recommended_price)
+            }, timeout=5)
+        except:
+            pass
+            
+    else:
+        score = analyzer.polarity_scores(book_title)['compound']
+        if score > 0.2:
+            result = f"Book not in database.\nSentiment: Positive (score: {round(score,3)})\nSuggestion: Consider maintaining or increasing price."
+        else:
+            result = f"Book not in database.\nSentiment: Neutral/Negative (score: {round(score,3)})\nSuggestion: Consider reviewing your pricing."
+        
+        # Send to n8n webhook
+        try:
+            requests.post(N8N_WEBHOOK_URL, json={
+                "book": book_title,
+                "sentiment": round(score, 3),
+                "recommendation": "UNKNOWN - not in database",
+                "current_price": None,
+                "recommended_price": None
+            }, timeout=5)
+        except:
+            pass
+    
+    return result
+
+iface = gr.Interface(
+    fn=analyze_book,
+    inputs=gr.Textbox(
+        label="Enter Book Title",
+        placeholder="e.g. A Light in the Attic"
+    ),
+    outputs=gr.Textbox(label="Analysis Results"),
+    title="📚 Book Price Optimizer",
+    description="Enter a book title to get sentiment analysis and AI-powered pricing recommendations. Results are automatically sent to our n8n automation pipeline.",
+    examples=[
+        ["A Light in the Attic"],
+        ["Tipping the Velvet"],
+        ["Sharp Objects"]
+    ]
+)
+
+iface.launch()