This course was created by Doug Rose. We are pleased to offer this training in our library.
- Define ethics and distinguish between virtue ethics, utilitarianism, and objectivism.
- Define and compare the categorical imperatives.
- Describe what it means for an algorithm to be traceable.
- Compare situations in which data should be accessible and inaccessible.
- Define data bias and describe how it can arise.
- Describe approaches to combat data bias and achieve fairness.
Skill Level Beginner
- Now, more than ever, computer systems are making complex decisions about you. You might have machine learning systems that make decisions about your job applicants or computer algorithms that makes decisions about your customers, but there are a lot of ethical challenges around these decisions. This course is designed for managers, directors, and developers that work with organizational data. It'll give you a framework for how to make decisions about data ethics in a typical office meeting. First, we'll talk about whether or not your decisions need to be traceable. Does your customer have a right to know exactly how a decision was made? Do you have an ethical responsibility to tell someone why the system denied them a loan or gave them a low rating? Then we'll look at the balance between objectivity and fairness when you see bias in your data. Should your organization change the data to make it fair based on income, race, or gender? It's a difficult balance between your business interests, fairness, and transparency. Algorithms can certainly help you make quick decisions, the challenge is making sure that these decisions reflect the best interest of your customer and your organizational values.