Name:
interface
Value:
Amplify has re-imagined the way frontend developers build fullstack applications. Develop and deploy without the hassle.

Choose your framework/language

Page updated Nov 21, 2024

Data Extraction

Data extraction allows you to parse unstructured text and extract structured data using AI. This is useful for converting free-form text into typed objects that can be used in your application.

The following example shows how to extract product details from an unstructured product description. The AI model will analyze the text and return a structured object containing the product name, summary, price, and category.

amplify/data/resource.ts
const schema = a.schema({
ProductDetails: a.customType({
name: a.string().required(),
summary: a.string().required(),
price: a.float().required(),
category: a.string().required(),
}),
extractProductDetails: a.generation({
aiModel: a.ai.model('Claude 3 Haiku'),
systemPrompt: 'Extract the property details from the text provided',
})
.arguments({
productDescription: a.string()
})
.returns(a.ref('ProductDetails'))
.authorization((allow) => allow.authenticated()),
});
src/components/Example.tsx
import type { Schema } from "../amplify/data/resource";
import { generateClient } from "aws-amplify/api";
import { createAIHooks } from "@aws-amplify/ui-react-ai";
const client = generateClient<Schema>({ authMode: "userPool" });
const { useAIGeneration } = createAIHooks(client);
export default function Example() {
const productDescription = `The NBA Official Game Basketball is a premium
regulation-size basketball crafted with genuine leather and featuring
official NBA specifications. This professional-grade ball offers superior grip
and durability, with deep channels and a moisture-wicking surface that ensures
consistent performance during intense game play. Priced at $159.99, this high-end
basketball belongs in our Professional Sports Equipment category and is the same model
used in NBA games.`
// data is React state and will be populated when the generation is returned
const [{ data, isLoading }, extractProductDetails] =
useAIGeneration("extractProductDetails");
const productDetails = async () => {
extractProductDetails({
productDescription
});
};
}