---
title: "Comment by George smith on \"How to make an app?\" | Hashnode"
description: "This is so interesting in the main building blocks of our dataset. You are called to ser features which can be categorical or numerical. But to work with algori"
url: "https://hashnode.com/posts/how-to-make-an-app-lessons-learned-after-running-a-software-company-for-13-years/5d95f389b6cff5785ccae4a7/comment/60a41b0a5c1fcb4077363ec5"
---

[Discussion](https://hashnode.com/posts/how-to-make-an-app-lessons-learned-after-running-a-software-company-for-13-years/5d95f389b6cff5785ccae4a7)

Comment [How to make an app?](https://hashnode.com/posts/how-to-make-an-app-lessons-learned-after-running-a-software-company-for-13-years/5d95f389b6cff5785ccae4a7)

[George smith](https://hashnode.com/@georgeone)

[May 18, 2021](https://hashnode.com/posts/how-to-make-an-app-lessons-learned-after-running-a-software-company-for-13-years/5d95f389b6cff5785ccae4a7/comment/60a41b0a5c1fcb4077363ec5)

This is so interesting in the main building blocks of our dataset. You are called to ser features which can be categorical or numerical. But to work with [algorithmic trading](https://www.altrady.com/blog/algorithmic-crypto-trading) you need an actual draft working. Simply put, categorical data into the group data with similar features while numerical data gives information with numbers of trading outcomes.
