论文信息 - Making intelligent systems understandable and controllable by end users

Making intelligent systems understandable and controllable by end users

Pervasive systems for end users are becoming mainstream yet ways to make them transparent and controllable by users are still in their infancy. In this position paper we describe our work with other kinds of intelligent systems to make them intelligible and adaptable by end users. Our results could hold useful lessons for pervasive systems to better support their use.

[1] Todd Kulesza,et al. Tell me more?: the effects of mental model soundness on personalizing an intelligent agent , 2012, CHI.

[2] Brad A. Myers,et al. Designing the whyline: a debugging interface for asking questions about program behavior , 2004, CHI.

[3] Thomas G. Dietterich,et al. Interacting meaningfully with machine learning systems: Three experiments , 2009, Int. J. Hum. Comput. Stud..

[4] Desney S. Tan,et al. Interactive optimization for steering machine classification , 2010, CHI.

[5] Weng-Keen Wong,et al. Integrating rich user feedback into intelligent user interfaces , 2008, IUI '08.

[6] Thomas G. Dietterich,et al. Toward harnessing user feedback for machine learning , 2007, IUI '07.

[7] Weng-Keen Wong,et al. End-user feature labeling: a locally-weighted regression approach , 2011, IUI '11.

[8] Anind K. Dey,et al. Why and why not explanations improve the intelligibility of context-aware intelligent systems , 2009, CHI.

[9] Weng-Keen Wong,et al. Why-oriented end-user debugging of naive Bayes text classification , 2011, ACM Trans. Interact. Intell. Syst..

[10] Weng-Keen Wong,et al. Explanatory Debugging: Supporting End-User Debugging of Machine-Learned Programs , 2010, VL/HCC.

[11] WongWeng-Keen,et al. Why-oriented end-user debugging of naive Bayes text classification , 2011 .

[12] Weng-Keen Wong,et al. Fixing the program my computer learned: barriers for end users, challenges for the machine , 2009, IUI.