Designing Alternative Representations of Confusion Matrices to Support Non-Expert Public Understanding of Algorithm Performance
The DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Abstract
Ensuring effective public understanding of algorithmic decisions that are powered by machine learning techniques has become an urgent task with the increasing deployment of AI systems into our society. In this work, we present a concrete step toward this goal by redesigning confusion matrices for binary classification to support non-experts in understanding the performance of machine learning models. Through interviews (n=7) and a survey (n=102), we mapped out two major sets of challenges lay people have in understanding standard confusion matrices: the general terminologies and the matrix design. We further identified three sub-challenges regarding the matrix design, namely, confusion about the direction of reading the data, layered relations and quantities involved. We then conducted an online experiment with 483 participants to evaluate how effective a series of alternative representations target each of those challenges in the context of an algorithm for making recidivism predictions. We developed three levels of questions to evaluate users' objective understanding. We assessed the effectiveness of our alternatives for accuracy in answering those questions, completion time, and subjective understanding. Our results suggest that (1) only by contextualizing terminologies can we significantly improve users' understanding and (2) flow charts, which help point out the direction of reading the data, were most useful in improving objective understanding. Our findings set the stage for developing more intuitive and generally understandable representations of the performance of machine learning models.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Using thematic analysis in psychology
Braun, V., & Clarke, V. (2006) · Qualitative Research in Psychology · doi:10.1191/1478088706qp063oa · Cited by 194,414
Explanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302
Identifying careless responses in survey data.
Meade, A. W., & Craig, S. B. (2012) · Psychological Methods · doi:10.1037/a0028085 · Cited by 3,865 · cited 1×
“We also included an instructed-response question for an attention check [37].”
MethodBarocas, S., & Selbst, A. D. (2014) · doi:10.2139/ssrn.2477899 · Cited by 3,048
Fair prediction with disparate impact: a study of bias in recidivism prediction instruments
Chouldechova, A. (2017) · Big Data · doi:10.1089/big.2016.0047 · Cited by 2,159 · cited 1×
“A concrete application scenario is to facilitate public understanding of the trade-offs embodied in different fairness metrics developed by the machine learning community [14, 29], for example, help lay people understand whether the underlying algorithm has a higher false positive rate in labeling black defendants tha…”
BackgroundLee, M. K. (2018) · Big Data & Society · doi:10.1177/2053951718756684 · Cited by 1,338
Amazon Scraps Secret AI Recruiting Tool that Showed Bias against Women *
Dastin, J. (2022) · Ethics of Data and Analytics · doi:10.1201/9781003278290-44 · Cited by 1,165
Inherent Trade-Offs in the Fair Determination of Risk Scores
Kleinberg, J., Mullainathan, S., & Raghavan, M. (2016) · arXiv · doi:10.48550/arxiv.1609.05807 · Cited by 941
Methods for Testing and Evaluating Survey Questionnaires
(2004) · doi:10.1002/0471654728 · Cited by 855
Making Sense of Graphs: Critical Factors Influencing Comprehension and Instructional Implications
Friel, S. N., Curcio, F. R., & Bright, G. W. (2001) · Journal for Research in Mathematics Education · doi:10.2307/749671 · Cited by 819 · cited 1×
“Our evaluation questions were developed and adapted from previous frameworks in learning science [20], which identified three levels of graph understanding of statistical results for lay people: Level 1 focuses on extracting or locating information; Level 2 focuses on finding relationships; and Level 3 focuses on anal…”
MethodDatta, A., Sen, S., & Zick, Y. (2016) · 2016 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2016.42 · Cited by 728 · cited 2×
“To date, the majority of past work on explaining and visualizing AI systems has primarily focused on supporting expert decision-making ([13, 16, 30, 31, 46]).”
Background'It's Reducing a Human Being to a Percentage': Perceptions of Justice in Algorithmic Decisions
Binns, R., Van Kleek, M., Veale, M., Lyngs, U., Zhao, J., & Shadbolt, N. (2018) · Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3173574.3173951 · Cited by 691
Methods for Testing and Evaluating Survey Questions
Presser, S., Couper, M. P., Lessler, J. T., Martin, E., Martin, J., Rothgeb, J. M., & Singer, E. (2004) · Public Opinion Quarterly · doi:10.1093/poq/nfh008 · Cited by 546
Fairness testing: testing software for discrimination
Galhotra, S., Brun, Y., & Meliou, A. (2017) · Proceedings of the 2017 11th Joint Meeting on Foundations of Software Engineering · doi:10.1145/3106237.3106277 · Cited by 405 · cited 1×
“…visualizations in (e.g., [28, 38]) used simple diagrams to target specific probability problems such as Bayesian Reasoning, the recent surge of XAI has led to more complex visual techniques, systems, and toolkits to facilitate expert users’ evaluation of model performance (e.g. see [3, 21]).”
MethodThe Principles and Limits of Algorithm-in-the-Loop Decision Making
Green, B., & Chen, Y. (2019) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3359152 · Cited by 392 · cited 1×
“, race) as targeting variables [44], which statistical definitions of fairness people perceive to be the fairest in the context of loan decisions [47], as well as how humans use AI systems to make decisions [23, 24].”
BackgroundExplaining Decision-Making Algorithms through UI: Strategies to Help Non-Expert Stakeholders
Cheng, H.-F., Wang, R., Zhang, Z., O'Connell, F., Gray, T., Harper, F. M., & Zhu, H. (2019) · Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3290605.3300789 · Cited by 292 · cited 4×
“Recent studies (e.g., [13, 35, 50, 56]) have started to use visualizations or create user interfaces to communicate algorithmic decisions to their study participants.”
BackgroundInfluentialSelf-citationThe Rise of Big Data Policing: Surveillance, Race, and the Future of Law Enforcement
Ferguson, A. G. (2020) · doi:10.2307/j.ctt1pwtb27 · Cited by 266
A Qualitative Exploration of Perceptions of Algorithmic Fairness
Woodruff, A., Fox, S. E., Rousso-Schindler, S., & Warshaw, J. (2018) · Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3173574.3174230 · Cited by 256
Value-Sensitive Algorithm Design: Method, Case Study, and Lessons
Zhu, H., Yu, B., Halfaker, A., & Terveen, L. (2018) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3274463 · Cited by 220
Self-citationBrown, A., Chouldechova, A., Putnam-Hornstein, E., Tobin, A., & Vaithianathan, R. (2019) · Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3290605.3300271 · Cited by 207
Human Decision Making with Machine Assistance: An Experiment on Bailing and Jailing
Grgić-Hlača, N., Engel, C., & Gummadi, K. P. (2019) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3359280 · Cited by 113 · cited 1×
“, race) as targeting variables [44], which statistical definitions of fairness people perceive to be the fairest in the context of loan decisions [47], as well as how humans use AI systems to make decisions [23, 24].”
BackgroundGrgić-Hlača, N., Redmiles, E. M., Gummadi, K. P., & Weller, A. (2018) · arXiv · doi:10.48550/arxiv.1802.09548 · Cited by 75
Cited by
The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice
Delgado, F., Yang, S., Madaio, M., & Yang, Q. (2023) · Equity and Access in Algorithms, Mechanisms, and Optimization · doi:10.1145/3617694.3623261 · Cited by 256
What is Human-Centered about Human-Centered AI? A Map of the Research Landscape
Capel, T., & Brereton, M. (2023) · Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3580959 · Cited by 228
“Design probes: [279, 303] Participatory design: [165] Field study: [114, 145, 148, 321] [26, 28, 44, 53, 62, 72, 103, 104, Design workshops: Co-design: [15, 19, 31, 117] Narrative study: [292] 111, 156, 162, 164, 175, 180, 187, [11, 216, 277, 282] Research through design: Mixed methods: 194, 231, 248, 253, 254, 261, 2…”
Seeing Like a Toolkit: How Toolkits Envision the Work of AI Ethics
Wong, R. Y., Madaio, M. A., & Merrill, N. (2023) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3579621 · Cited by 135
Zhang, S., Yu, J., Xu, X., Yin, C., Lu, Y., Yao, B., Tory, M., Padilla, L. M., Caterino, J., Zhang, P., et al. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642343 · Cited by 99
“In healthcare, there are a number of decision-making tasks that have similar properties as sepsis diagnosis: highly uncertain, high-stakes, and time-sensitive [41].”
BackgroundEffect of Information Presentation on Fairness Perceptions of Machine Learning Predictors
van Berkel, N., Goncalves, J., Russo, D., Hosio, S., & Skov, M. B. (2021) · Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445365 · Cited by 84
“Shen et al. explored alternative designs of confusion matrices to support people’s understanding of an algorithm’s performance, ultimately identifying flow charts as the most useful due to their ability to point out the direction of reading [49].”
Background
Metadata
- Authors
- Shen, H., Jin, H., Cabrera, Á. A., Perer, A., Zhu, H., & Hong, J. I.
- Venue
- Proceedings of the ACM on Human-Computer Interaction
- Published
- 2020-10-14
- Volume
- 4
- Issue
- CSCW2
- Pages
- 1-22
- ISSN
- 2573-0142 (ISSN-L)
- OpenAlex
- W3094294413
- Cited by
- 52
- References
- 22
- FWCI
- 2.84
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.