Explainable artificial intelligence practices can support data scientists in interpreting the results of machine learning (ML) models. However, current practices require effortful and time-consuming coding to compare explanations that either relate to different ML models of the same data, or that are generated with different computational methods. We report the development of a glyph-based polar chart (GPC), designed to support a more comprehensive interpretation of the results of ML models by allowing these comparisons. The results from our user experience evaluation indicated that the proposed GPC supported data scientists in identifying the most relevant model variables, comparing different explanation methods, and performing logic reviews.
BanieckiH.KretowiczW.PiatyszekP.WisniewskiJ.BiecekP. (2021). Dalex: Responsible machine learning with interactive explainability and fairness in python. Journal of Machine Learning Research, 22(214), 1–7. https://jmlr.csail.mit.edu/papers/volume22/20-1473/20-1473.pdf
3.
BengfortB.BilbroR. (2019). Yellowbrick: Visualizing the scikit-learn model selection process. The Journal of Open Source Software, 4(35), 1075. https://doi.org/10.21105/joss.01075
4.
BhattU.XiangA.SharmaS.WellerA.TalyA.JiaY.GhoshJ.PuriR.MouraJ. M. F.EckersleyP. (2020). Explainable machine Learning in deployment. Conference on fairness, accountability, and transparency, IL, USA, June 27 - 30, 2023, (pp. 1–10). https://doi.org/10.1145/3351095.3375624
5.
Doshi-VelezF.KimB. (2017). Towards a rigorous science of interpretable machine learning. arXiv. arXiv preprint arXiv:1702.08608.
6.
DruckerS. M.FernandezR. (2015). A unifying framework for animated and interactive unit visualizations. Microsoft Research. Technical Report MSR-TR-2015–65.
7.
GashiM.VukovićM.JekicN.ThalmannS.HolzingerA.Jean-QuartierC.JeanquartierF. (2022). State-of-the-Art explainability methods with focus on visual analytics showcased by glioma classification. BioMedInformatics, 2(1), 139–158. https://doi.org/10.3390/biomedinformatics2010009
8.
HelwigN.PignanelliE.SchützeA. (2015). Condition monitoring of a complex hydraulic system using multivariate statistics. Proceedings of IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Pisa, Italy, 16-19 May 2022, 210–215.
9.
HohmanF.HeadA.CaruanaR.DeLineR.DruckerS. M. (2019). Gamut: A design probe to understand how data scientists understand machine learning models proceedings of the CHI conference on human factors in computing systems (CHI ’19). ACM, 579(01), 1–13. https://doi.org/10.1145/3290605.3300809
10.
HongS. R.HullmanJ.BertiniE. (2020). Human factors in model interpretability: Industry practices, challenges, and needs. Proceedings of the ACM on Human-Computer Interaction, 4(CSCW1), 1–26. https://doi.org/10.1145/3392878
11.
JesusS.BelmC.BalayanV.BentoJ.SaleiroP.BizarroP.GamaJ. (2021). How can i choose an explainer?: An application-grounded evaluation of post-hoc explanations. Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, Virtual Event Canada (pp. 805–815). https://doi.org/10.1145/3442188.3445941
12.
KammerD.KeckM.GrunderT.MaaschA.ThomT.KleinsteuberM.GrohR. (2020). Glyphboard: Visual exploration of high-dimensional data combining glyphs with dimensionality reduction. IEEE Transactions on Visualization and Computer Graphics, 26(4), 1661–1671. https://doi.org/10.1109/TVCG.2020.2969060
13.
KaurH.NoriH.JenkinsS.CaruanaR.WallachH.Wortman VaughanJ. (2020). Interpreting interpretability: Understanding data scientists’ use of interpretability tools for machine learning. Proceedings of Conference on Human Factors in Computing Systems, Honolulu, HI, USA, 1–14. https://doi.org/10.1145/3313831.3376219
14.
KrishnaS.HanT.GuA.PombraJ.JabbariS.WuZ. S.LakkarajuH. (2022). The disagreement problem in explainable machine learning: A practitioner’s perspective. arXiv. https://arxiv.org/abs/2202.01602
15.
KumarP.SharmaM. (2021). Feature-importance feature-interactions (FIFI) graph: A graph-based novel visualization for interpretable machine learning. International Conference on Intelligent Technologies (CONIT), Hubli, India, 1–7. https://doi.org/10.1109/CONIT51480.2021.9498467
16.
KuzluM.CaliU.SharmaV.GulerO. (2020). Gaining insight into solar photovoltaic power generation forecasting utilizing explainable artificial intelligence tools. IEEE Access, 8, 187814–187823. https://doi.org/10.1109/ACCESS.2020.3031477
17.
LangerM.OsterD.SpeithT.HermannsH.KästnerL.SchmidtE.SesingA.BaumK. (2021). What do we want from explainable artificial intelligence (XAI)? – a stakeholder perspective on XAI and a conceptual model guiding interdisciplinary XAI research. Artificial Intelligence, 296, 103473. https://doi.org/10.1016/j.artint.2021.103473
18.
LeeD. J. L.TangD.AgarwalK.BoonmarkT.ChenC.KangJ.MukhopadhyayU.SongJ.YongM.HearstM. A.ParameswaranA. G. (2021). Lux: Always-on visualization recommendations for exploratory dataframe workflows. Proceedings of the VLDB Endowment, 15(3), 727–738. https://doi.org/10.14778/3494124.3494151
19.
LiaoQ. V.GruenD.MillerS. (2020). Questioning the AI: Informing design practices for explainable AI user experiences. Proceedings of the 2020 CHI conference on human factors in computing systems, Honolulu, HI, USA (pp. 1–15). https://doi.org/10.1145/3313831.3376590
20.
LundbergS. M.LeeS. (2017). A unified approach to interpreting model predictions. Proceedings of the 31st international conference on neural information processing systems, Long Beach, CA, USA (pp. 4768–4777). https://dl.acm.org/doi/10.5555/3295222.3295230
21.
ParkD.DruckerS. M.FernandezR.ElmqvistN. (2018). Atom: A grammar for unit visualizations. IEEE Transactions on Visualization and Computer Graphics, 24(12), 3032–3043. https://doi.org/10.1109/TVCG.2017.2785807
22.
ZhouJ.HuangW.ChenF. (2020). A radial visualisation for model comparison and feature identification. IEEE Pacific Visualization Symposium, Tianjin, China, 226–230. https://doi.org/10.1109/PacificVis48177.2020.1028
