Data science (DS) projects often follow a lifecycle that consists of laborious tasks for data scientists and domain experts (e.g., data exploration, model training, etc.). Only till recently, machine learning(ML) researchers have developed promising automation techniques to aid data workers in these tasks. This paper introduces Au-toDS, an automated machine learning (AutoML) system that aims to leverage the latest ML automation techniques to support data science projects. Data workers only need to upload their dataset, then the system can automatically suggest ML configurations, preprocess data, select algorithm, and train the model. These suggestions are presented to the user via a web-based graphical user interface and a notebook-based programming user interface. We studied Au-toDS with 30 professional data scientists, where one group used AutoDS, and the other did not, to complete a data science project. As expected, AutoDS improves productivity; Yet surprisingly, we find that the models produced by the AutoDS group have higher quality and less errors, but lower human confidence scores. We reflect on the findings by presenting design implications for incorporating automation techniques into human work in the data science lifecycle.
CCS CONCEPTS• Human-centered computing → User studies; Empirical studies in HCI ; • Computing methodologies → Artificial intelligence.
Video chat systems such as Skype, Google+ Hangouts, and FaceTime have been widely adopted by family members and friends to connect with one another over distance. We have conducted a corpus of studies that explore how various demographics make use of such video chat systems where this usage moves beyond the paradigm of conversational support to one in which aspects of everyday life are shared over long periods of time, sometimes in an almost passive manner. We describe and reflect on studies of longdistance couples, teenagers, and major life events, along with design research focused on new video communication systems-the Family Window, Family Portals, and Perch-that explicitly support 'alwayson video' for awareness and communication. Overall, our findings show that people highly value long-term video connections and have appropriated them in a number of different ways. Designers of future video communication systems need to consider: ways of supporting the sharing of everyday life, rather than just conversation; providing different design solutions for different locations and situations; providing appropriate audio control and feedback; and, supporting expressions of intimacy over distance.
Much ICTD research for sub-Saharan Africa has focused on how technology related interventions have aimed to incorporate marginalized communities towards global economic growth. Our work builds on this. We present results from an exploratory qualitative study on the family communication practices of family members who communicate both within and between rural, suburban, and urban settings in Kenya. Our findings reveal that family communication focuses on economic support, well-being, life advice, and everyday coordination of activities. We also outline social factors that affect family communication, including being an eldest child, having a widowed sibling, and having reduced access to technology because of gender, literacy, or one's financial situation. Lastly, we discuss new opportunities for technology design and articulate the challenges that designers will face if creating or deploying family communication technologies in Kenya.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.