A scalable decision-tree-based method to explain interactions in dyadic data

Abstract

Gaining relevant insight from a dyadic dataset, which describes interactions between two entities, is an open problem that has sparked the interest of researchers and industry data scientists alike. However, the existing methods have poor explainability, a quality that is becoming essential in certain applications. We describe an explainable and scalable method that, operating on dyadic datasets, obtains an easily interpretable high-level summary of the relationship between entities. To do this, we propose a quality measure, which can be configured to a level that suits the user, that factors in the explainability of the model. We report experiments that confirm better results for the proposed method over alternatives, in terms of both explainability and accuracy. We also analyse the method's capacity to extract relevant actionable information and to handle large datasets.

Explainable Artificial Intelligence: Objectives, Stakeholders, and Future Research Opportunities

“…Besides only very few examples (e.g. Eiras-Franco et al, 2019;Giboney et al, 2015;Martens & Provost, 2014), since then most of the publications 1 on explainability of AI systems, or "Explainable Artificial Intelligence" (XAI), have been published outside of the information systems community, mostly in computer science. As one can see, the existing IS literature is very valuable but with its peak in the 1990ies and early 2000s also comparatively dated, which motivates our call for more IS research on the explainability of AI.…”

Regression Tree Based Explanation for Anomaly Detection Algorithm

“…The model used to obtain the pre-hoc explanation will consist of a grouping of the input patterns attending to their numerical variables. Clusters will be defined as the leaf nodes of a shallow decision tree [6]. Each pattern will be assigned its ADMNC estimator [5].…”

“…where NV(Cl i ) represents the number of variables needed to describe cluster Cl i and λ is a hyperparameter that allows the supervisor to balance the accuracy and interpretability [6] of the whole clustering. This quality measure is always negative and the goal of the algorithm is maximizing its value to approach 0.…”

Explainable artificial intelligence: a comprehensive review

“…Decision tree learning, which is a commonly used hierarchical structure-based ML algorithm for the regression and the classification topics (Eiras-Franco et al 2019;Gupta et al 2017;Wang et al 2019d), can be interpreted easily and fulfill all the constraints of the inherently transparent model. It has consistently stayed between various groups of transparent models.…”