ABSTRACT
This chapter discusses data products, the requirement to build pipelines and processes and the need for creating machine learning models able to perform scoring on device. It provides an example for deploying a simple machine learning model in a mobile device such as an Apple iPhone. With the high availability of a large number of connected devices, the computing power that can be harnessed for the application of machine learning is huge. There are some tools that enable the deployment of machine learning models into a device, taking a trained model and encapsulating it in a format that is compatible with the ecosystem of the device in question. The chapter discusses the end-to-end creation of a simple app with machine learning at its core. It explores the complexity of data products: Ranging from raw data and algorithms through to automated decision making.
