A framework for user assistance on predictive models - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

A framework for user assistance on predictive models

Résumé

Data analysis generally requires very specialized skills, especiallywhen applying machine learning tasks. The ambition of the paperis to propose a framework assisting a domain expert user to analysehis data, in a context of predictive analysis. In particular, the frame-work includes a recommender system for the workflow of analysistasks. Because the lack of explanation in recommendations can leadto loss of confidence, a complementary system is proposed to betterunderstand the predictive models recommended. This complemen-tary system aims to help the user to understand and exploit theresults of the data analysis, by relying on his data expertise. Theframework is validated through a pool of questions and a mock-upshowing the interest of the approach.
Fichier principal
Vignette du fichier
Circle_2020.pdf (1.59 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03138319 , version 1 (11-02-2021)

Licence

Paternité - Pas d'utilisation commerciale - Pas de modification

Identifiants

  • HAL Id : hal-03138319 , version 1

Citer

Gabriel Ferrettini, Julien Aligon, Chantal Soulé-Dupuy, William Raynaut. A framework for user assistance on predictive models. Joint Conference of the Information Retrieval Communities in Europe (CIRCLE 2020), Jul 2020, Samatan, Gers, France. pp.32. ⟨hal-03138319⟩
66 Consultations
36 Téléchargements

Partager

Gmail Facebook X LinkedIn More