On The Role of Knowledge Graphs in Explainable AI

Submitted by Freddy Lecue on 07/19/2019 - 08:30

Tracking #: 2259-3472

Authors: 
Freddy Lecue

Responsible editor: 
Guest Editor 10-years SWJ

Submission type: 
Other
Abstract: 
The current hype of Artificial Intelligence (AI) mostly refers to the success of machine learning and its sub-domain of deep learning. However, AI is also about other areas, such as Knowledge Representation and Reasoning, or Distributed AI, i.e., areas that need to be combined to reach the level of intelligence initially envisioned in the 1950s. Explainable AI (XAI) now refers to the core backup for industry to apply AI in products at scale, particularly for industries operating with critical systems. This paper reviews XAI not only from a Machine Learning perspective, but also from the other AI research areas, such as AI Planning or Constraint Satisfaction and Search.We expose the XAI challenges of AI fields, their existing approaches, limitations and opportunities for Knowledge Graphs and their underlying technologies.
Full PDF Version: 
PDF icon swj2259.pdf
Previous Version: 
On The Role of Knowledge Graphs in Explainable AI
Tags: 
Reviewed

Decision/Status: 
Accept

Solicited Reviews:
Click to Expand/Collapse
Review #1
Anonymous submitted on 05/Aug/2019
Suggestion:
Accept
Review Comment:

This paper has addressed reviewers' comments and has improved both in clarity and in description of opportunities. Only a few very minor typographic defects exist that need to be addressed, relating to references. These include dangling reference [?], overlapping texts, and unnecessary boldfacing in references. Overall this paper is a nice overview of the burgeoning area of XAI.

Review #2
By Dagmar Gromann submitted on 08/Aug/2019
Suggestion:
Accept
Review Comment:

Thanks for the detailed revision of the paper and for addressing all issues raised.

I thought the reference:
Pomarlan, M., Porzel, R., Bateman, J., & Malaka, R. (2018, November). From sensors to sense:
Integrated heterogeneous ontologies for Natural Language Generation. In Proceedings of the Workshop
on NLG for Human–Robot Interaction (pp. 17-21).

could be a good fit for Section 2.7. since it proposes a reason-aloud approach, but feel free to ignore if you do not think it is relevant for this section.

The potential co-vision paper "Neural Language Models for the Multilingual, Transcultural, and Multimodal Semantic
Web" could only be addressed in terms of further challenges of NLP and Section 2.10, but probably not such a good fit. Feel free to omit.

It seems like not all minor comments have been addressed (e.g. Neural Netwok => Network) - please make sure all spotted errors and typos are addressed.

Review #3
By Guilin Qi submitted on 13/Sep/2019
Suggestion:
Minor Revision
Review Comment:

The authors have addressed my comments. I think the submission can be accepted if the authors can have a proof-reading of the paper. Some references are missing and we can find [?] in the paper. Some sentences are hard to parse, such as "linking knowledge graphs extracts" and "where explanations could be elaborated their a latent representations"