h(odor): Interactive Discovery of Hypotheses on the Structure-Odor Relationship in Neuroscience

Guillaume Bosc (1) , Marc Plantevit (1) , Jean-François Boulicaut (1) , Moustafa Bensafi (2) , Mehdi Kaytoue (1)

1 DM2L - Data Mining and Machine Learning
2 CRNL - Centre de recherche en neurosciences de Lyon - Lyon Neuroscience Research Center

From a molecule to the brain perception, olfaction is a complex phenomenon that remains to be fully understood in neuroscience. Latest studies reveal that the physico-chemical properties of volatile molecules can partly explain the odor perception. Neuroscientists are then looking for new hypotheses to guide their research: physico-chemical descriptors distinguishing a subset of perceived odors. To answer this problem, we present the platform h(odor) that implements descriptive rule discovery algorithms suited for this task. Most importantly, the ol-faction experts can interact with the discovery algorithm to guide the search in a huge description space w.r.t their non-formalized background knowledge thanks to an ergonomic user interface.

Mots clés: pattern mining; interactive; subgroup discovery

Domaines: Informatique [cs]; Algorithme et structure de données [cs.DS]; Base de données [cs.DB]

Guillaume Bosc, Marc Plantevit, Jean-François Boulicaut, Moustafa Bensafi, Mehdi Kaytoue. h(odor): Interactive Discovery of Hypotheses on the Structure-Odor Relationship in Neuroscience. ECML/PKDD 2016 (Demo), Sep 2016, Riva del Garda, Italy. pp.17--21. ⟨hal-01346679⟩