Transparency in machine learning modeling
Transparency helps users of your model trust it by helping them understand how it works and how it was built. It also helps you, your team, your collaborators, and your organization to collect feedback on different components of your machine learning life cycle. It is worth understanding the transparency requirements in different stages of a life cycle and the challenges in achieving them:
- Data collection: Transparency in data collection needs to answer two major questions:
- What data are you collecting?
- What do you want to use that data for?
For example, when users click on the agreement button for data usage when registering for a mobile phone app, they are giving consent for the information they provide in the app to be used. But the agreement needs to be clear on the part of the user data that is going to be used and for what purposes.
- Data selection and exploration: In these stages of the life cycle, your process of...
