Abstract

Abstract
The internet is used by the majority of the world's population. Many of its contents are free for consumers, supported by digital marketing investment. The current large online population can render digital marketing campaigns inefficient for brands buying ads if the right message is not reaching the right audience. Customer Data Platforms (CDPs) enables brands to collect first-party data about their customers and leverage it to reach the right audience, without sharing private customer data with third parties. This paper documents how CDPs work, by detailing their main components as patterns and relating them as a pattern language. The language is composed of five patterns: Event Tracking, ID Matching, User Profile Storage, Segmentation, and Activation. The pattern language can be used by marketeers and engineers in digital marketing as an introduction or reference to how CDPs are designed and work. © 2022 ACM.

Abstract
Proliferation of the Internet is enabling the use of sensors and actuators to capture data and control devices remotely in a multitude of domains. Still, there is a general lack of best practices while designing such large scale real-time systems. This paper describes a generic architecture used on the implementation of a framework for deploying such systems in the cloud, enabling run-time evolution of the system with new sensors, actuators or services possibly developed by third-parties being integrated dynamically. Such architecture orchestrates the flow of information in the ecosystem and scales transparently to external components when needed, requiring no change in them. Adoption in the Portuguese nation-wide AAL project AAL4ALL is then described. Copyright © 2013 by the Association for Computing Machinery, Inc. (ACM).