Arguments in Interactive Machine Learning

In most applications of machine learning, domain experts provide domain speciñc knowledge. From previous experience it is known that domain experts are unable to provide all relevant knowledge in advance, but need to see some results of machine learning hrst. Interactive machine learning, where expe...

Full description

Saved in:
Bibliographic Details
Published inInformatica (Ljubljana) Vol. 42; no. 1; pp. 53 - 59
Main Author Možina, Martin
Format Journal Article
LanguageEnglish
Published Ljubljana Slovenian Society Informatika / Slovensko drustvo Informatika 01.03.2018
Subjects
Algorithms
Artificial intelligence
Data mining
Debugging
Experts
International conferences
Knowledge acquisition
Knowledge discovery
Machine learning
Subject specialists
New York
Online AccessGet full text

Cover

More Information
Summary:In most applications of machine learning, domain experts provide domain speciñc knowledge. From previous experience it is known that domain experts are unable to provide all relevant knowledge in advance, but need to see some results of machine learning hrst. Interactive machine learning, where experts and machine learning algorithm improve the model in turns, seems to solve this problem. In this position paper, we propose to use arguments in interaction between machine learning and experts. Since using and understanding arguments is a practical skill that humans learn in everyday life, we believe that arguments will help experts to better understand the models, facilitate easier elicitation of new knowledge from experts, and can be intuitively integrated in machine learning. We describe an argument-based dialogue, which is based on a series of steps such as questions and arguments, that can help obtain from a domain expert exactly that knowledge which is missing in the current model.
ISSN:0350-5596
1854-3871