ML & GenAI

"Generate a summary of customer feedback trends"
"Create a product description based on technical specifications"
"Summarize this SQL query results: [data]"

-- First, execute a query
"Execute 'SELECT product, COUNT(*) as issues FROM support_tickets GROUP BY product' on my_connection"

-- Then generate a summary
"Generate a summary of the product support ticket distribution"

"Generate a welcome email for new customers mentioning our key features"
"Create a FAQ answer explaining our refund policy"
"Write a brief explanation of database normalization"

"Generate insights from our quarterly sales data"
"Analyze the error patterns in system logs and suggest improvements"
"Summarize customer sentiment from review comments"

"What is the average delay incurred by flights?"
"Show me the top 5 customers by total order value"
"How many support tickets were opened last month?"

"What is the average order value by customer segment?"
"Show monthly revenue trends for the last year"
"Which products have the highest return rate?"

"How many active users do we have?"
"List all orders pending shipment"
"Show me employees hired in the last 6 months"

"Compare sales performance between regions"
"What's the difference in customer satisfaction between product categories?"
"Show the top and bottom performing stores"

SELECT AVG(arrival_delay) as avg_delay
FROM flights
WHERE arrival_delay IS NOT NULL;

SELECT customer_id, customer_name, SUM(order_total) as total_value
FROM orders
JOIN customers ON orders.customer_id = customers.id
GROUP BY customer_id, customer_name
ORDER BY total_value DESC
LIMIT 5;

"Which tables contain employee salary information?"
"Show me the schema for customer-related tables"
"What tables have pricing data?"

CREATE TABLE employees (
  id INT PRIMARY KEY,
  first_name VARCHAR(50),
  last_name VARCHAR(50),
  email VARCHAR(100),
  department_id INT,
  hire_date DATE,
  salary DECIMAL(10,2)
);

CREATE TABLE departments (
  id INT PRIMARY KEY,
  name VARCHAR(100),
  manager_id INT
);

CREATE TABLE employee_performance (
  employee_id INT,
  review_date DATE,
  rating INT,
  comments TEXT
);

"What tables exist for order processing?"
"Show me all tables related to inventory management"

"Which tables reference the customers table?"
"Show me the schema for the sales data model"

"What tables do I need to join to get customer purchase history?"
"Show me tables with product pricing information"

"How do I train a model in HeatWave AutoML?"
"What machine learning algorithms does HeatWave support?"
"How do I use ML_PREDICT_ROW?"

"What are the steps to train a machine learning model?"
"How do I prepare data for AutoML?"
"What's the difference between ML_TRAIN and ML_MODEL_LOAD?"

"What parameters can I use with ML_TRAIN?"
"How do I specify the target column for prediction?"
"Can I train classification and regression models?"

"How do I make predictions on new data?"
"What's the syntax for ML_PREDICT_TABLE?"
"How do I interpret prediction results?"

"How do I list all trained models?"
"Can I export a trained model?"
"How do I delete an old model?"

# Step 1: Discover relevant tables
"Which tables contain sales data?"

# Step 2: Ask analytical questions
"What is the average order value by region?"

# Step 3: Generate insights
"Generate a summary of regional sales performance trends"

# Step 4: Create embeddings for text search (if needed)
"Add embeddings for 'product_description' column into 'description_vec' column in products table"

# Step 5: Perform semantic search
"Search products for items similar to 'wireless headphones with noise cancellation'"

# Analyze customer behavior
"Show me customer segments by purchase frequency"

# Generate insights
"Generate insights about customer churn patterns from the segment data"

# Natural language queries
"Which customers haven't purchased in the last 90 days?"
"What products do high-value customers prefer?"

# Query ticket data
"How many support tickets were resolved last week?"

# Embed ticket descriptions
"Add embeddings for 'description' column into 'description_vec' column in tickets table"

# Semantic search for similar issues
"Search tickets for issues similar to 'login timeout error'"

# Generate summary
"Generate a summary of common support issues from last month"