Explainable Structuring and Discovery of Relevant Cases for Exploration of High-Dimensional Data - Université de Reims Champagne-Ardenne Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Explainable Structuring and Discovery of Relevant Cases for Exploration of High-Dimensional Data

Résumé

Data described by numerous features create a challenge for domain experts as it is difficult to manipulate, explore and visualize them. With the increased number of features, a phenomenon called "curse of dimensionality" arises: sparsity increases and distance metrics are less relevant as most elements of the dataset become equidistant. The result is a loss of efficiency for traditional machine learning algorithms. Moreover, many state-of-the-art approaches act as black-boxes from a user point of view and are unable to provide explanations for their results. We propose an instance-based method to structure datasets around important elements called exemplars. The similarity measure used by our approach is less sensitive to high-dimensional spaces, and provides both explainable and interpretable results: important properties for decision-making tools such as recommender systems. The described algorithm relies on exemplar theory to provide a data exploration tool suited to the reasoning used by experts of various fields. We apply our method to synthetic as well as real-world datasets and compare the results to recommendations made using a nearest neighbor approach.
Fichier principal
Vignette du fichier
ESIDA19_paper_4.pdf (1.21 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02088283 , version 1 (02-04-2019)

Identifiants

  • HAL Id : hal-02088283 , version 1

Citer

Joris Falip, Fréderic Blanchard, Michel Herbin. Explainable Structuring and Discovery of Relevant Cases for Exploration of High-Dimensional Data. Intelligent User Interfaces (IUI) Workshops, 2019, Los Angeles, United States. ⟨hal-02088283⟩

Collections

URCA CRESTIC
116 Consultations
58 Téléchargements

Partager

Gmail Facebook X LinkedIn More