FlavorMiner: a machine learning platform for extracting molecular flavor profiles from structural data
Fabio Herrera-Rocha, Miguel Fernández-Niño, Jorge Duitama, Mónica P Cala, María José Chica, Ludger A Wessjohann, Mehdi D Davari, Andrés Fernando González Barrios · 2024
Read FlavorMiner: a machine learning platform for extracting molecular flavor profiles from structural data by Fabio Herrera-Rocha, Miguel Fernández-Niño, Jorge Duitama, Mónica P Cala, María José Chica, Ludger A Wessjohann, Mehdi D Davari, Andrés Fernando González Barrios (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11633011.txt.
Description
Taste is a molecule's shadow, and labs spend dear to trace which bits cast which notes. Fabio Herrera-Rocha and seven co-authors built a platform in 2024 that guesses a molecule's flavor notes from its shape alone. Their tool, FlavorMiner, tries many pairings of learning method and shape code, with tricks to keep rare notes from being swamped. Two pairings, a forest and a neighbor vote over rich shape codes, led the field.
On the whole the guesses scored high on the standard curve mark. Turned on cocoa's small-stuff map, the tool pulled out notes and links that hand work would take months to find. Training spanned more than nine hundred foods, so the guesses travel. The team offer it to food labs drowning in spectra.
Catalog record
| CREATOR | Fabio Herrera-Rocha, Miguel Fernández-Niño, Jorge Duitama, Mónica P Cala, María José Chica, Ludger A Wessjohann, Mehdi D Davari, Andrés Fernando González Barrios |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |