Intelligent notification systems are the key building blocks of modern computing systems. Indeed, from mobile phones to desktop computers and from personal assistants to wearable devices, notifications are becoming a key part of our interactions with interactive systems. With the continuous integration of machine learning and artificial intelligence algorithms in digital systems, the underlying components for deciding what, when, and where to send notifications are becoming more and more intelligent. This book provides an in-depth introduction to the state of the art of this fascinating field. It is aimed at both researchers and practitioners working in this area or interested in exploring this highly cross-disciplinary topic.
The book only assumes a very basic knowledge of human-computer interaction and ubiquitous computing concepts. In general, several definitions are provided and extensive bibliography can be found at the end of book. The structure and contents of this book can be summarized as follows.
• In Chapter 1 (Introduction), we introduce the scope of the book and its aims, outlining the key issues in designing intelligent notification systems.
• In Chapter 2 (Understanding Users’ Interaction with Notifications), we frame the problem by examining possible definitions of interruptions based on past and current theories, give an overview of their types and discuss human response to such interruptions.
• In Chapter 3 (Costs Associated to Interruptions), we present the findings of various studies that examined the cost associated with the arrival of interruptions at inopportune moments. More specifically, we discuss the detrimental effects of interruptions on users’ memory, emotional and affective states, and ongoing task execution.
• In Chapter 4 (Personalization of Interruptibility Strategies), we provide an overview of the findings of studies concerning personalization of interruptibility strategies. We focus in particular on the effectiveness of personalization strategies with respect to time allocation in a multitasking environment. Moreover, we report the findings of studies demonstrating that the performance in carrying out an interrupted task is affected by users’ anxiety and arousal levels.
• In Chapter 5 (Design Principles for Interruptibility Management Systems), we analyze the design of interruption management systems and discuss the difference between users’ attentiveness and receptivity to interruptions.
• In Chapter 6 (Interruptibility Management Systems for Desktop Environments), we first introduce the characteristics of interruptions in desktop environment as they arrive in a constant context (i.e., desktops are not taken around with users). We then discuss the studies in the area of interruptibility management for desktop environment that were conducted by using a Wizard of Oz approach, by exploiting task phases and through on-the-fly inference of interruptibility.
• In Chapter 7 (Interruptibility Management Systems for Mobile Environments), we first discuss how the advent of mobile phones has provided opportunities for users to connect to different information channels and receive updates in real time about a variety of events. We then discuss the studies in the area of interruptibility management for mobile environment based on the analysis of current activity, the transition between activities, contextual data, and those concerning filtering of irrelevant information.
• In Chapter 8 (Open Challenges and Outlook), we first provide an overview of the limitations of existing studies that focus on understanding and learning users’ behavior in terms of interactions with notifications. We then summarize the key elements of this lecture outlining the open challenges in the field.
We hope that this book will stimulate further developments in this field. In the end, the acceptance of Artificial Intelligence technologies also depends on the design of efficient and effective intelligent solutions for the delivery of information to the end-users. Notifications are an essential class of interaction modality for intelligent technologies. For this reason, we believe a strong foundation in this area is important for everyone involved in the design of such systems.
Abhinav Mehrotra and Mirco Musolesi
December 2019