A Practical Application of Black Box Interpretability

Abstract In a world environment where technology is advancing at a rapid pace, fluent communication between machines and humans is becoming increasingly relevant. Interpretability through summarisation and explainability are imperative factors in the progression of machine and human communication. Using IBM Watson's visual recognition service, in conjunction with other processes we explore and establish interpretability. Specifically, we looked at identifying cancerous cells from biopsy images through visual analytics. Furthermore, future practical applications, including in different fields, were also investigated, i.e. counter IED operations.
  • Byungwoo (Peter) Jang
  • Mingu (Jason) Jeong
  • Hevin Na
  • Conner Russell
  • Dave Braines (IBM UK)
  • Richard Tomsett (IBM UK)
  • Graham White (IBM UK)
  • Dan Harborne (Cardiff)
Date Sep-2017
Venue 1st Annual Fall Meeting of the DAIS ITA, 2017