The ML project life cycle

A typical ML project life cycle starts by understanding the problem at hand. Typically, someone in the organization (possibly a data scientist or business stakeholder) feels that some part of their business can be improved by the use of ML. For example, a music streaming company could conjecture that providing recommendations of songs similar to those played by a user would improve user engagement with the platform. Once we understand the business context and possible business actions to take, the data science team will need to consider several aspects during the project life cycle.

The following diagram describes various steps in the ML project life cycle: