Introduction to AI in Business

First, we'll start by teaching you about the foundations of AI and machine learning: where is it used in industry and how do these technologies work? This section will discuss the type of data you need to create an AI product, the business cases that stand to benefit the most from AI-medicated technology, and the qualities of a good AI product team.

Creating a Dataset

In this section, you'll learn how to create your own, novel dataset using Figure Eight's data annotation platform. Data annotation is all about structuring your data such that a machine learning model can learn to automatically find patterns within that data. Here, you'll learn the best design practices for creating a dataset of your own.

Build a Model

In this section, you'll see how to build and train and end-to-end deep learning model to recognize patterns in a medical image dataset. You'll look at metrics that define the success of your trained model and parameters that affect how it trains.

Measuring Business Impact

In this final section, you'll learn how to measure the efficacy of your model after it is released. This section discusses methods for identifying bias, updating a model in response to underlying changes in the data, and end-to-end case studies that demonstrate how AI products are ever improving and evolving.

Course Projects

Intremittently, at the end of certain courses, is a project lesson. Each of these three projects must be completed to successfully complete this program. A completed project looks like the following image.

While incomplete projects will look like the following image.

Project-based Learning

At Udacity, we believe in learning by doing. So, after learning the theory behind AI for business products—through a series of videos, text, and exercises—you'll be tasked with applying your knowledge to projects.

Projects are, typically, practical writing or technical exercises that allow you to demonstrate some industry-relevant skills.
They are designed with the help of industry partners, so we make sure that we are teaching you the most relevant skills for becoming an AI product owner or manager!
When you complete each project, you will submit it to one of our project reviewers, who will give you personalized feedback on your submission.

Completing the projects will not only help you build your skills with what is taught in the lessons, but also show you how those skills are used in practice and build out your technical portfolio.

Don't worry if you are not familiar with how you would even approach some of the items discussed below, you will be learning the needed skills in the lessons ahead!

Project 1: Create a Medical Image Annotation Job

In this project, you will design a data annotation job for a dataset of chest xray images; some of which contain signs of pneumonia. Your goal, as a product owner is to build a product that helps doctors quickly identify cases of pneumonia. As such, this project is designed to test your ability to build a labeled dataset that distinguishes between healthy and pneumonia x-ray images.

A labeled, healthy, chest x-ray image.

Project 2: Build a Model

Take the next step and build a complete classification model that would serve as the backbone for a medical image classification product. (Don't worry, there's no coding involved!) For this task, you will use Google AutoML, an automated machine learning tool that will allow you to build the model and host it in the cloud. You'll measure the performance of multiple different models and evaluate them.

Results of a trained model using AutoML

Project 3: Create an AI Product Business Proposal

Put everything together in this final project; develop a complete business proposal for an AI product. The proposal will think through data considerations, evaluation metrics for a trained model, and design best practices for developing an AI product.

These projects are meant to test your ability to apply what you've learned throughout this course. Give yourself time to work on these projects, iterate and improve based on any reviewer feedback you get), and develop something that you can show off in a technical portfolio of work. You're encouraged to reach out to Udacity on Twitter or create a blog post of your own to share what you've learned!