Exploring eating behaviours modelling for user clustering

Sema Akkoyunlu (1) , Cristina Manfredotti (1) , Antoine Cornuéjols (1, 2) , Nicolas Darcel (3) , Fabien Delaere (4)

1 MIA-Paris - Mathématiques et Informatique Appliquées
2 MIA Paris-Saclay - Mathématiques et Informatique Appliquées
3 PNCA - Physiologie de la Nutrition et du Comportement Alimentaire
4 Danone Research

Food based dietary guidelines are not fully adopted by consumers. One of the principal explanations for this failure is that they are too general and do not take into account eating habits. Experts in nutrition believe that providing personalized dietary recommendations via nutrition recommender system can help people improve their eating behaviours. Understanding eating habits is a keystone in order to build a context aware recommender system that delivers personalized dietary recommendations. As a step towards this goal, we propose a method for representing food consumptions based on Doc2Vec for discovering clusters of eating behaviours. We compare our method to the state of the art methods used in the nutrition community.

Mots clés: CCS CONCEPTS • Information systems → Information extraction; • Human- centered computing → User models; KEYWORDS food recommender systems; user modelling; eating behaviours; Doc2Vec

Domaines: Apprentissage [cs.LG]; Alimentation et Nutrition

Sema Akkoyunlu, Cristina Manfredotti, Antoine Cornuéjols, Nicolas Darcel, Fabien Delaere. Exploring eating behaviours modelling for user clustering. HealthRecSys@RecSys 2018 colocated with ACM Recsys’18 (ACM Conference Series on Recommender Systems), Oct 2018, Vancouver, Canada. pp.46-51. ⟨hal-02480562⟩