Although machine learning is becoming commonly used in today's software, there has been little research into how end users might interact with machine learning systems, beyond communicating simple "right/wrong" judgments. If the users themselves could somehow work hand-in-hand with machine learning systems, the accuracy of learning systems could be improved...
This paper presents qualitative results from interviews with knowledge workers about their recovery strategies after interruptions. Special focus is given to when these strategies fail due to the nature of the interruption and existing computer support. Potential solutions offered by participants to overcome some of these problems are presented. These...