User-oriented Assessment of Classification Model Understandability

By Hiva Allahyari, Niklas Lavesson. 2011

71 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Systems

Citations in the corpus

Or's of And's for Interpretable Classification, with Application to Context-Aware Recommender Systems (2015, background): “Both models have priors that can be adjusted to suit a domain-specific notion of interpretability, as it is well-known that interpretability comes in different forms for different domains [3, 15, 21, 29, 30, 39].”
Methods and Models for Interpretable Linear Classification (2014, neutral): “This is consistent with field studies on interpretability, which often conclude that different predictive models are “most” interpretable in different domains for different reasons [103, 60, 1, 51].”
How do Humans Understand Explanations from Machine Learning Systems? An Evaluation of the Human-Interpretability of Explanation (2018, background): “, 2003] and the kind that increase the odds of a good decision [Bilgic and Mooney, 2005]. In many cases, these studies are looking at whether the explanation has an effect, sometimes also considering a few different kinds of explanation (actions of similar customers, etc.). Horsky et al. [2012] desc”
Explaining prediction models and individual predictions with feature contributions (2014, background): “Other studies focused only on decision trees [20] and on decision trees and decision rules [2].”
Explainable and Interpretable Models in Computer Vision and Machine Learning (2017, background): “For example, in many studies, the comprehensibility of learned rules is assumed to be negatively correlated with their complexity, a point of view that has been questioned in recent work (Allahyari and Lavesson 2011; Stecher et al. 2016; Fürnkranz et al. 2018).”
An Evaluation of the Human-Interpretability of Explanation (2019, background): “Allahyari and Lavesson [2011], Elomaa [2017] also find that larger models can sometimes be more interpretable. Schmid et al. [2016] find that human-recognizable intermediate predicates in inductive knowledge programs can sometimes improve simulation time.”
Comprehensibility of Classification Trees–Survey Design (2019, background): “The first task - classify asks respondent to classify an instance according to a given classifier (same as in [1, 12]).”
Explainable Artificial Intelligence: a Systematic Review (2020, methodology): “The surveys carried out in [342] aimed at comparing six supervised learning algorithms.”
Comprehensible classification models: a position paper (2014, methodology): “First, in some studies users found larger models to be more comprehensible than smaller models [1], [35], [46], apparently because larger models provided more classificationrelevant information to users.”
A comprehensive taxonomy for explainable artificial intelligence: a systematic survey of surveys on methods and concepts (2021, background): “decision trees as empirically evaluated in [66,49,4];”
A Bayesian Framework for Learning Rule Sets for Interpretable Classification (2017, background): “Both can be adjusted to suit a domain-specific notion of interpretability; it is well-known that interpretability comes in different forms for different domains (Martens and Baesens, 2010; Martens et al., 2011; Huysmans et al., 2011; Allahyari and Lavesson, 2011; Rüping, 2006; Freitas, 2014).”
Human Evaluation of Models Built for Interpretability (2019, background): “Allahyari and Lavesson (2011); Elomaa (2017) also ﬁnd that larger models can sometimes be more interpretable.”
RuleMatrix: Visualizing and Understanding Classifiers with Rules (2018, result): “[15] found that decision tables are the most effective representations, while other studies [2] disagrees.”
Improving the interpretability of classification rules discovered by an ant colony algorithm (2013, background): “In addition, there is evidence showing that, in some applications, larger models were considered by users as more comprehensible than smaller ones, since they contained more informative attributes to the user [11, 10].”
Interpretability of machine learning models and representations: an introduction (2016, methodology): “Allahyari and Lavesson used a survey filled by users to evaluate the interpretability of models generated by 6 learning algorithms [26].”