Explore

Extract data from a page with the help of AI

⚠️ Test your scenarios and save them at every step!

You are the head of sales at an impact-focused company. You want to work with like-minded entrepreneurs. To reach that goal, you are going to scrape the website of the Earthshot Prize, a prize awarded to ventures and initiatives that work towards preserving the climate and the planet.

🟢 Level 1 - Filter out irrelevant data

1️⃣Create a Make Scenario and rename it.

2️⃣ Add an “HTTP - Make a request” module to your scenario

3️⃣Run the “HTTP - Make a request” module to retrieve the content of this page :

https://earthshotprize.org/winners-finalists/re-green/⁠

. You don’t need to add anything in the body of the request.

4️⃣Remove the HTML tags from the page you just retrieved.

Clue : You will need to use a module from the “Text parser” app in Make.

5️⃣Extract only the text that contains information that is relevant to the

re.green⁠

venture.

Clue :

You will need to set a variable using the “Tools - Set Variable” to create and store this new text.

You can use the

split()⁠

and

get()⁠

functions to break the text into chunks and return the one you are interested in. Click on the links to learn how they work !

6️⃣Test your scenario with other finalist pages from the same website.

🟠 Level 2 - Use AI to extract structured data

1️⃣Add an “OpenAI - Generate a response” module to your scenario. Use the same settings as those you have used in

2️⃣Configure the OpenAI module the same way you did in this practice exercise :

No Code Course 2 - Albert School · Play around with the Open AI API⁠

⁠

3️⃣The OpenAI module must output a structured JSON containing the following information : • Type (Winner or Finalist) • Year • Earthshot category • Project name • Project summary (three sentences) • Area of impact • Contact name • YouTube page URL • LinkedIn page URL • Instagram page URL 💡You can use a JSON formatter like

https://jsoneditoronline.org/⁠

to ensure your JSON is well formatted!

4️⃣Test your scenario to make sure the OpenAI module outputs only JSON.

5️⃣Add a “JSON - Parse JSON” module to your scenario.

6️⃣Create an Airtable base and name it Earthshot Finalists.

7️⃣Create a table that can store the data extracted from the page.

8️⃣ Add an “Airtable - Create a record” module to your scenario

9️⃣ Save the scrapped data in the Airtable base you’ve just created.

🔟 Test your scenario with other finalist pages from the same website.

Want to print your doc?
This is not the way.

Try clicking the ··· in the right corner or using a keyboard shortcut (

CtrlP

) instead.