"Guidance on Instructing the Model: Defining SQL Join Operations for Text-to-SQL Query Generation"

I am converting natural language to SQL queries using GPT-3.5-Turbo. I have provided the model with a JSON-structured schema containing table names, column names, and descriptions for each column in the tables.

During the conversion process, there are instances where the model selects incorrect tables or columns when performing JOIN operations. I am seeking guidance on how to externally teach the language model about JOIN operations more effectively. What would be the most effective method to enhance the model’s understanding in this specific aspect?"

What’s your current prompt? You could be more specific and describe the tables and column names so that the generated SQL query is correct.