Pierre Le Bras
Texture Lab / Strategic Futures Laboratory,
School of Mathematical and Computer Science,
Heriot-Watt University, Edinburgh, EH14 4AS
pl196 at hw dot ac dot uk
+44 (0)131 451 4166
LinkedIn profile, Website
- Improving User Confidence in Concept Maps: Exploring Data Driven Explanations
- Pierre Le Bras, David A. Robb, Thomas S. Methven, Stefano Padilla, Mike J. Chantler
- In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems. ACM, 2018.
- Developing and Debugging Proof Strategies by Tinkering
- Yuhui Lin, Pierre Le Bras, and Gudmund Grov
- In International Conference on Tools and Algorithms for the Construction and Analysis of Systems, pp. 573-579. Springer Berlin Heidelberg, 2016.
My research interests are a mixture of Data Science, Human-Computer Interaction, and Machine Learning. In particular, I investigates user confidence and trust in automated systems, i.e. explainable AI. My Ph.D. research focuses on user confidence in automated topic map visualisation. In particular, I have implemented and analysed the effects of data-driven explanations when used to improve user confidence. My investigations led me to draw out confidence-driven usability recommendations.
Using the Topic Hex Map application I developed for my research, we produced an overview of British and European funded research in the UK.
Website : strategicfutures.org/hexmaps/ukeutopics/2017-01/
I have been involved in the Tinker project since my last year of undergraduate studies.
Tinker is a tool enabling Proof Strategies to be represented as graphs (called PSGraphs) and works on top of Theorem Provers (e.g. Isabelle, ProofPower).
My main work was to re-create an interface that would allow more functionalities with PSGraphs such as hierarchies and a better communication protocol between the interface and the Theorem Provers.
Website : ggrov.github.io/tinker