Thesis Explainable AI: Explainable Configuration with a Constraint-based Product Configurator
Zeitpunkt der Veröffentlichung: 08.04.2021
Teaser: Would you like to combine research and industry in your thesis? Then join our Future Labs team and work on the latest IT trends.
During the configuration process there are two ways how product attributes can be included in the configuration: either, attributes are selected by the user (we call these decisions the user wish), or they are derived by the configurator because they are implied by the user wish and the rules of the product. Since the rules of a product may be complex and plentiful it is often difficult to understand for the user why the configurator selected certain attributes. The configurator can already provide explanations for its changes, however, these explanations are in the form of an implication trace which is difficult to understand for most users. The challenge of this thesis is to find a way how to present these explanations in a user friendly way.
· Design and implement a REST service that can provide the explanation of a specific configuration change
· Find a way to convert implication traces into user friendly explanations
· Modify the frontend applications (Angular) so that they can display these explanations upon the user's request
· Summarize the results in your thesis and/or scientific publication(s)
· Jack of all trades: can do frontend with Angular and backend with JAVA
· Ability to empathize with tech-noobs
· Creativity in approaching hard problems
· Communicating your ideas clearly and precisely in English
. Working together with others in German is also easy for you
Interested? Then email us your CV and latest transcript of records at email@example.com