SAsSy - Making Decisions Transparent with Argumentation and Natural Language Generation

Nava Tintarev, Roman Kutlak

Research output: Contribution to conferenceUnpublished paperpeer-review

15 Downloads (Pure)

Abstract

An autonomous system consists of one or more physical or virtual agents that can perform tasks without continuous human guidance. In order to realise their promise, techniques for making such autonomous systems scrutable and transparent are therefore required. To address this issue the Scrutable Autonomous Systems (SAsSy) demo shows how argumentation and natural language can be combined to generate a human understandable dialog explaining the operation of an autonomous system. On the one hand argumentation theory is used to simulate human understandable reasoning mechanisms. On the other, natural language generation tools are used to translate logical statements into simple plain English. The idea is to generate a dialog that enables the user to understand and question the reasoning present in autonomous systems.
Original languageEnglish
Publication statusPublished - 24 Feb 2014
EventIUI 2014 Workshop on Interacting with Smart Objects - Haifa, Israel
Duration: 24 Feb 201424 Feb 2014

Workshop

WorkshopIUI 2014 Workshop on Interacting with Smart Objects
Country/TerritoryIsrael
CityHaifa
Period24/02/1424/02/14

Keywords

  • explanations
  • argumentation
  • natural language generation
  • agents

Fingerprint

Dive into the research topics of 'SAsSy - Making Decisions Transparent with Argumentation and Natural Language Generation'. Together they form a unique fingerprint.

Cite this