Fixing the program my computer learned: barriers for end users, challenges for the machine

The results of a machine learning from user behavior can be thought of as a program, and like all programs, it may need to be debugged. Providing ways for the user to debug it matters, because without the ability to fix errors users may find that the learned program's errors are too damaging for them to be able to trust such programs. We present a new approach to enable end users to debug a learned program. We then use an early prototype of our new approach to conduct a formative study to determine where and when debugging issues arise, both in general and also separately for males and females. The results suggest opportunities to make machine-learned programs more effective tools.

[1]  A. Bandura.  Self-efficacy: toward a unifying theory of behavioral change. , 1977, Psychological review.

[2]  Deborah L. McGuinness,et al.  Toward establishing trust in adaptive agents , 2008, IUI '08.

[3]  Gregg Rothermel,et al.  End-user software engineering with assertions in the spreadsheet paradigm , 2003, 25th International Conference on Software Engineering, 2003. Proceedings..

[4]  Brad A. Myers,et al.  Getting more out of programming-by-demonstration , 1999, CHI '99.

[5]  Henry Lieberman,et al.  Supporting User Hypotheses in Problem Diagnosis on the Web and Elsewhere , 2004 .

[6]  Henry Lieberman,et al.  Supporting user hypotheses in problem diagnosis , 2004, IUI '04.

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

[8]  Duane Szafron,et al.  Visual Explanation of Evidence with Additive Classifiers , 2006, AAAI.

[9]  Margaret M. Burnett,et al.  Testing vs. code inspection vs. what else?: male and female end users' debugging strategies , 2008, CHI.

[10]  Carmen Lacave,et al.  A review of explanation methods for Bayesian networks , 2002, The Knowledge Engineering Review.

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

[12]  Peter Norvig,et al.  Artificial Intelligence: A Modern Approach , 1995 .

[13]  Brad A. Myers,et al.  Demonstrational and constraint-based techniques for pictorially specifying application objects and behaviors , 1995, TCHI.

[14]  Andreas Wierse,et al.  Information Visualization in Data Mining and Knowledge Discovery , 2001 .

[15]  A. Adam.  Whatever happened to information systems ethics? Caught between the devil and the deep blue sea , 2004 .

[16]  Eser Kandogan,et al.  Koala: capture, share, automate, personalize business processes on the web , 2007, CHI.

[17]  James A. Landay,et al.  Investigating statistical machine learning as a tool for software development , 2008, CHI.

[18]  Brad A. Myers,et al.  Six Learning Barriers in End-User Programming Systems , 2004, 2004 IEEE Symposium on Visual Languages - Human Centric Computing.

[19]  Igor Kononenko,et al.  Inductive and Bayesian learning in medical diagnosis , 1993, Appl. Artif. Intell..

[20]  Deborah Compeau,et al.  Application of Social Cognitive Theory to Training for Computer Skills , 1995, Inf. Syst. Res..

[21]  Brad A. Myers,et al.  Answering why and why not questions in user interfaces , 2006, CHI.

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

[23]  Jiun-Hung Chen,et al.  Recovering from errors during programming by demonstration , 2008, IUI '08.

[24]  Ron Kohavi,et al.  Visualizing the Simple Bayesian Classi er , 1997 .

[25]  Brad A. Myers,et al.  Asking and answering questions about the causes of software behavior , 2008 .

[26]  Henry Lieberman,et al.  Your Wish is My Command , 2001, The Morgan Kaufmann series in interactive technologies.

[27]  Todd Kulesza,et al.  Can feature design reduce the gender gap in end-user software development environments? , 2008, 2008 IEEE Symposium on Visual Languages and Human-Centric Computing.

[28]  Adnan Darwiche,et al.  When do Numbers Really Matter? , 2001, UAI.

[29]  Margaret M. Burnett,et al.  Effectiveness of end-user debugging software features: are there gender issues? , 2005, CHI.