Explainable AI represents an increasingly important category of systems that attempt to support human understanding and trust in machine intelligence and automation. Typical systems rely on algorithms to help understand underlying information about decisions and establish justified trust and reliance. Researchers have proposed using goodness criteria to measure the quality of explanations as a formative evaluation of an XAI system, but these criteria have not been systematically investigated in the literature. To explore this, we present a novel collaborative explanation system (CXAI) and propose several goodness criteria to evaluate the quality of its explanations. Results suggest that the explanations provided by this system are typically correct, informative, written in understandable ways, and focus on explanation of larger scale data patterns than are typically generated by algorithmic XAI systems. Implications for how these criteria may be applied to other XAI systems are discussed.
KincaidJ. P.FishburneR. P.JrRogersR. L.ChissomB. S. (1975). Derivation of new readability formulas (automated readability index, fog count and flesch reading ease formula) for navy enlisted personnel. Naval Technical Training Command Millington TN Research Branch.
5.
LimB. Y.DeyA. K.AvrahamiD. (2009). Why and why not explanations improve the intelligibility of context-aware intelligent systems. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 2119–2128.
6.
McHughM. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 22(3), 276–282.
7.
MuellerS. T.VeinottE. S.HoffmanR. R.KleinG.AlamL.MamunT.ClanceyW. J. (2021). Principles of Explanation in Human-AI Systems. ArXiv Preprint ArXiv:2102.04972.
8.
OhS. (2018). Social Q&A. In BrusilovskyP.HeD. (Eds.), Social Information Access: Systems and Technologies (pp. 75–107). Springer International Publishing. https://doi.org/10.1007/978-3319-90092-6_3
9.
PapenmeierA.EnglebienneG.SeifertC. (2019). How model accuracy and explanation fidelity influence user trust. ArXiv Preprint ArXiv:1907.12652.
MamunT.HoffmanR. R.MuellerS.T. (2021, July). Collaborative Explainable AI: A non-algorithmic approach to generating explanations of AI. In International Conference on Human-Computer Interaction. Springer.
12.
Doshi-VelezF.KimB. (2017). A Roadmap for a Rigorous Science of Interpretability. ArXiv Preprint ArXiv:1702.08608. https://arxiv.org/abs/1702.08608
HoffmanR. R.MuellerS. T.KleinG.LitmanJ. (2018a). Metrics for explainable AI: Challenges and prospects. ArXiv Preprint ArXiv:1812.04608.
15.
HoffmanR. R.KleinG.MuellerS. T. (2018b). Explaining explanation for “explainable ai.”Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 62(1), 197–201.
16.
HuangG.FangC. H.AgarwalN.BhagatN.EloyJ. A.LangerP. D. (2015). Assessment of online patient education materials from major ophthalmologic associations. JAMA Ophthalmology, 133(4), 449–454.
17.
KleinG.HoffmanR.MuellerS. (2019). Naturalistic Psychological Model of Explanatory Reasoning: How people explain things to others and to themselves. International Conference on Naturalistic Decision Making.
18.
KuleszaT.BurnettM.WongW.-K.StumpfS. (2015). Principles of explanatory debugging to personalize interactive machine learning. Proceedings of the 20th International Conference on Intelligent User Interfaces, 126–137. https://doi.org/10.1145/2678025.2701399
19.
MuellerS. T.HoffmanR. R.ClanceyW.EmreyA.KleinG. (2019). Explanation in human-AI systems: A literature meta-review, synopsis of key ideas and publications, and bibliography for explainable AI. ArXiv Preprint ArXiv:1902.01876.
20.
SwartoutW. R.MooreJ. D. (1993). Explanation in second generation expert systems. Second Generation Expert Systems, 543, 585.
21.
WickM. R.SlagleJ. (1989). The partitioned support network for expert system justification. IEEE Transactions on Systems, Man, and Cybernetics, 19(3), 528–535. https://doi.org/10.1109/21.31059
22.
WoolfB. (2007). Building Intelligent Interactive Tutors: Student-centered strategies for revolutionizing e-learning. Morgan Kaufmann Publishers Inc.
23.
MuellerS. T.AgarwalP.LinjaA.DaveN.AlamL. (2020, December). The Unreasonable Ineptitude of Deep Image Classification Networks. In Proceedings of the Human Factors and Ergonomics Society Annual Meeting (Vol. 64, No. 1, pp. 410414). Sage CA: Los Angeles, CA: SAGE Publications.
