Designing Interaction Around Smart and Autonomous Systems

计算能力的进步和更复杂的机器学习算法，以及数字宇宙规模的翻倍，正在导致一个日益自治的世界。尽管这提供了帮助我们做出更好决策、更有效地工作和生活的机会，但目前尚不清楚我们如何围绕复杂的ML和AI算法设计人类交互，这些算法不仅对最终用户来说难以理解，对交互产品的设计师来说也是如此，以实现真正的智能和自主现实。这个项目旨在通过使用一个应用程序来提高我们对人类与更复杂和智能的自主系统互动的理解，该应用程序可以自动将通知分为不同的类别，根据用户当前所处的活动类型选择只显示某些通知。工作通知在办公时间和休闲通知在晚上。这项研究将遵循既定的方法，包括在野外研究中，用户在他们的手机上安装应用程序并设置通知配置，然后他们的应用程序使用情况将被监控，以探索不同用户的互动水平和信任或依赖程度，以及这种情况如何随着时间的推移而变化。

Advancements in computational power and more complex machine learning algorithms, along with a doubling in the size of the digital universe are leading to an increasingly autonomous world. Although this presents opportunities to help us make better decisions, work and live more efficiently, it is unclear as yet how we can design human interaction around complex ML and AI algortihms that are often impenetrable not only for end users, but also for the designers of interactive products in order to achieve a truly smart and autonomous reality. This project aims to advance our understanding of human interaction with more complex and smart autonomous systems by using an app that automates the process of separating notifications into different categories, choosing to display only certain notifications depending on the type of activity the user is currently in, eg. work notifications during office hours and leisure notifications at night. The research will follow established methods involving in the wild studies where the user installs the app on their phone and sets up the notification configuration - their app usage will then be monitored to explore the level of interaction and trust or reliance that different users have and how this may change over time.