Application of Machine Learning for E-justice

Metsker, Oleg G.; Trofimov, Egor Viktorovich; Kopanitsa, Georgy

doi:10.1088/1742-6596/1828/1/012006

Cited by 4 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The importance of transparency, explanations and interpretations of machine learning models is growing, particularly for decision making in high risk and safety critical applications (Kim et al 2016), including for example clinical decision support systems (Antoniadi et al 2021) for example for cancer detection, distinguishing between fraudulent and genuine claims to an insurance company (Rawat et al 2021), autonomous navigation systems supervised by humans (Brandsaeter et al 2020), or decision support systems in law enforcement intended to improve legal practice (Metsker et al 2021). Ribeiro et al (2016) claim that "if the users do not trust a model or a prediction, they will not use it".…”

Section: Introductionmentioning

confidence: 99%

Shapley values for cluster importance

Brandsæter

Glad

2022

Data Min Knowl Disc

View full text Add to dashboard Cite

This paper proposes a novel approach to explain the predictions made by data-driven methods. Since such predictions rely heavily on the data used for training, explanations that convey information about how the training data affects the predictions are useful. The paper proposes a novel approach to quantify how different data-clusters of the training data affect a prediction. The quantification is based on Shapley values, a concept which originates from coalitional game theory, developed to fairly distribute the payout among a set of cooperating players. A player’s Shapley value is a measure of that player’s contribution. Shapley values are often used to quantify feature importance, ie. how features affect a prediction. This paper extends this to cluster importance, letting clusters of the training data act as players in a game where the predictions are the payouts. The novel methodology proposed in this paper lets us explore and investigate how different clusters of the training data affect the predictions made by any black-box model, allowing new aspects of the reasoning and inner workings of a prediction model to be conveyed to the users. The methodology is fundamentally different from existing explanation methods, providing insight which would not be available otherwise, and should complement existing explanation methods, including explanations based on feature importance.

show abstract