Harnessing graph neural networks to craft fragrances based on consumer feedback

Rodrigues, Bruno C.L.; Santana, Vinicius V.; Queiroz, Luana P.; Rebello, Carine M.; B․ R․ Nogueira, Idelfonso

doi:10.1016/j.compchemeng.2024.108674

Cited by 2 publications

(1 citation statement)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By implementing this approach, the number of molecules to be considered is reduced, as it suggests fewer, more suitable options, reducing the time of development of a new fragrance substantially. 47 However, the composition is not determined. Hence, it still needs to be tested and optimized through trial and error.…”

Section: Introductionmentioning

confidence: 99%

Molecule Generation and Optimization for Efficient Fragrance Creation

Rodrigues,

Santana,

Murins

et al. 2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

This research introduces a machine learning-centric approach to replicate olfactory experiences, utilizing an experimentally quantified target scent from the literature as a case study for validation. Key contributions encompass a hybrid model connecting perfume molecular structure to human olfactory perception. This model includes an AI-driven molecule generator (utilizing graph and generative neural networks), quantification and prediction of odor intensity, and refinement of optimal solvent and molecule combinations for desired fragrances. Additionally, a thermodynamically based model establishes a link between olfactory perception and liquid-phase concentrations. The methodology employs transfer learning and selects the most suitable molecules based on vapor pressure and fragrance notes. Ultimately, a mathematical optimization problem is formulated to minimize discrepancies between new and target olfactory experiences. The methodology is validated by reproducing two distinct olfactory experiences using available experimental data.

show abstract

Section: Introductionmentioning

confidence: 99%