Affective Crowdsourcing Applied to Usability Testing

Usability tests are very important to validate and improve software systems and software development processes. Nowadays, there are different methodologies to software usability evaluation. However, the most effective are based on real user's feedback although these methodologies are time-consuming and very expensive because depend on analysis of many human computer interactions. In this paper two hypotheses are launched. First, usability tests could be applied remotely based on crowdsourcing platforms. Second, outliers could be detected based on user's emotional behaviour. In order to investigate these hypotheses an affective usability evaluation process has been developed and supported by a low cost software interface called Hesitation Detector (HD). An HD is an affective component programmed to recognize events related to user's emotional states. In this paper a deterministic automata is presented with purpose of automatic detection of hesitation, which qualifies it to be applied in different computational platforms, including tablets and smartphones. Besides, it not requires obstructive sensors because user’s movements are gathered and processed by means traditional interfaces. In order to validate and customize the proposed automata some usability tests were performed together with biomedical signals processing. Preliminary results reveal evidences that affective crowdsourcing can be a promising alternative in order to evaluated users satisfaction and their reactions remotely across the Internet. Keywords-Affective Systems; Affective Crowdsourcing; Usability Testing.

[1]  Arjen P. de Vries,et al.  Obtaining High-Quality Relevance Judgments Using Crowdsourcing , 2012, IEEE Internet Computing.

[2]  Arturas Kaklauskas,et al.  Biometric Mouse Intelligent System for Student's Emotional and Examination Process Analysis , 2009, 2009 Ninth IEEE International Conference on Advanced Learning Technologies.

[3]  Judith J. Azcarraga,et al.  Use of Personality Profile in Predicting Academic Emotion Based on Brainwaves Signals and Mouse Behavior , 2011, 2011 Third International Conference on Knowledge and Systems Engineering.

[4]  John Riedl,et al.  Crowdsourcing medical research , 2013, Computer.

[5]  Landon P. Cox.  Truth in Crowdsourcing , 2011, IEEE Security & Privacy.

[6]  Dana Chisnell.  Usability testing: Taking the experience into account , 2010, IEEE Instrumentation & Measurement Magazine.

[7]  Robert Morris,et al.  Crowdsourcing Workshop: The Emergence of Affective Crowdsourcing , 2011 .

[8]  Gonzalo Bailador,et al.  A Stress-Detection System Based on Physiological Signals and Fuzzy Logic , 2011, IEEE Transactions on Industrial Electronics.

[9]  Yuan Qi,et al.  Context-sensitive Bayesian classifiers and application to mouse pressure pattern classification , 2002, Object recognition supported by user interaction for service robots.

[10]  M. Hossain,et al.  Crowdsourcing: Activities, incentives and users' motivations to participate , 2012, 2012 International Conference on Innovation Management and Technology Research.

[11]  Elizabeth Gerber,et al.  Affect and Creative Performance on Crowdsourcing Platforms , 2013, 2013 Humaine Association Conference on Affective Computing and Intelligent Interaction.

[12]  M. Tarvainen,et al.  Principal component analysis of galvanic skin responses , 2000, Proceedings of the 22nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society (Cat. No.00CH37143).

[13]  Rosalind W. Picard.  Affective Computing , 1997 .