> ## Documentation Index
> Fetch the complete documentation index at: https://docs.trynota.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Data Generation

> How to create workflow prompts for generating and uploading synthetic data files.

## Example Prompts

You can use natural language or structured lists in your workflow prompt.\
For example:

* Upload a CSV file with columns `["name", "email", "city"]`
* Upload a JSON file with fields `{"student_id", "dob", "grade"}`
* Upload a file with `['employee_name', 'position', 'salary']`
* Upload a CSV with headers `["first_name", "last_name", "email", "phone_number"]`
* Upload a file where each record has `{ "product", "price", "category" }`
* Upload a JSON file using `{ "customer_name", "order_id", "amount" }`
* Upload a file with fields `[ 'user_id', 'signup_date', 'referrer' ]`
* Upload a structured file with keys `{ "id", "name", "registered_on" }`
* Upload a file with format `[ 'doctor_name', 'department', 'appointment_date' ]`

NOTA parses your prompt and generates a matching synthetic dataset.

***

## CSV Input Format

To generate a **CSV**, include a list of column headers in your prompt.\
For example:

```txt theme={null}
["Student ID", "Full Name", "Date of Birth", "Gender", "Grade Level", "Email"]
```

NOTA automatically maps these headers to suitable data generators, such as:

```txt theme={null}
["uuid4", "name", "date_of_birth", "random_gender", "random_grade", "email"]
```

Your output will be a CSV file with the columns you requested, each filled with relevant synthetic data.

> **Tip:** You can provide column names in any format—brackets, quotes, or plain text.

***

## JSON Input Format

For **JSON** files, provide the list of field names in your prompt.\
For example:

```txt theme={null}
{"Student ID", "Full Name", "Date of Birth", "Gender"}
```

NOTA will generate sample JSON records such as:

```json theme={null}
{
  "Student ID": "7481023",
  "Full Name": "Alice Johnson",
  "Date of Birth": "1994-11-27",
  "Gender": "Female"
}
```

***

## No Structure? No Problem.

If you don’t specify a structure, NOTA will generate a file with default sample fields and data—just ask to "create and upload a test file" in your workflow prompt.

***

## How NOTA Interprets Your Fields

NOTA uses language understanding to match your field names to the most relevant generator, even if it isn’t an exact match.\
For example:

* `mobile` → phone number
* `dob` → date of birth
* `zipcode` → postal code

Just describe your data—NOTA handles the mapping.

***

## Summary Table

| Format | Input Example       | Output Example                          |
| ------ | ------------------- | --------------------------------------- |
| CSV    | List of headers     | CSV with synthetic rows and columns     |
| JSON   | List of field names | JSON records with realistic sample data |

***

Need help? Describe your data in plain language, be specific and NOTA will take care of the rest.