• Corpus ID: 35950405

Using Watson for Constructing Cognitive Assistants

@inproceedings{Goel2016UsingWF,
  title={Using Watson for Constructing Cognitive Assistants},
  author={Ashok K. Goel and Jordan Belknap and Brian Creeden and Mithun Kumble and Shanu Salunke and Bradley Sheneman and Bryan Wiltgen},
  year={2016}
}

• Ashok K. Goel, J. Belknap, +4 authors Bryan Wiltgen
• Published 2016
• Computer Science

We describe an experiment in using IBM’s Watson tool to construct cognitive assistants in a Georgia Tech Spring 2015 class on computational creativity. The project-based class used Watson to support biologically inspired design, a paradigm that uses biological systems as analogues for inventing technological systems. The students worked in small teams and developed semester-long projects that used Watson to build cognitive assistants for conducting research in support of biologically inspired design. In this paper, we describe this experiment in using Watson, present two of the projects in detail, draw best practices for using Watson, and characterize and critique the Watson tool for constructing cognitive assistants. 1. Background and Motivations In 2011 IBM’s Watson system surprised and delighted much of the world by winning the game show Jeopardy! against human opponents. Since then IBM has claimed Watson to be a “cognitive system” that has ushered in an era of “cognitive computing” (Kelly & Hamm 2013). It has published several articles on Watson (e.g., Brown et al. 2013; Ferruci et al. 2010; Kalyanpur & Murdock 2015), including a special issue of the IBM Journal of Research and Development (e.g., Ferruci 2012). In addition, it has incrementally provided access to various versions of Watson, starting with the skeletal Watson Engagement Advisor (n.d.) that is freely available to most faculty and students for educational purposes, and recently, the more robust Bluemix Platform (n.d.) that provides access to various components of Watson as services and is available to most anyone for most any purpose but typically at a charge. Given the large-scale IBM publicity and investment, Watson has come to exemplify cognitive systems in much of the popular media. Yet, at least outside IBM, our understanding of Watson remains modest. The articles published by IBM often go into enormous algorithmic and representational detail on selected technical issues, but typically leave out both critical elements of the representations and algorithms as well