Machine learning for smell: Ordinal odor strength prediction of molecular perfumery components

2025/12/09 by Peter Fichtelmann, Julia Westermayr, Fichtelmann, Peter +1 · 1 voice
Agricultural and Biological Sciences · Engineering · Neuroscience · Physics and Astronomy · #Advanced Chemical Sensor Technologies #Chemical Physics (physics.chem-ph) #FOS: Physical sciences #Insect Pheromone Research and Control #Olfactory and Sensory Function Studies #physics.chem-ph

paper · pdf · doi:10.48550/arxiv.2512.08683

openalex publication_date 2025/12/09 · arxiv published 2025/12/09 · arxiv updated 2025/12/09 · openalex created_date 2025/12/11 · openalex updated_date 2026/07/28

Abstract

Predicting olfactory perception directly from molecular structure is central to fragrance design that plays a role in a wide range of industries, such as perfumery, food and beverage, and health care. Among olfactory attributes, odor strength is a key factor in shaping odor perception, but its modeling has been impeded by scarce and fragmented intensity data. In this work, we introduce an ordinal odor strength data set of over 2,000 molecules by integrating two different public sources, mapping structures to odorless, low, medium, and high categories. Across several molecular encodings and supervised learning algorithms we compared different prediction strategies. Dimensionality reduction and SHAP analysis identifies molecular size, polarity, ring features, and branching as primary drivers, consistent with mass-transport constraints on volatility, sorption, and receptor access. This scalable ordinal framework enables reliable odor-strength estimation for novel molecules and provides a foundation for in silico fragrance design.

Citations

• Optuna: A Next-generation Hyperparameter Optimization Framework
• Explaining individual predictions when features are dependent: More accurate approximations to Shapley values
• Rank consistent ordinal regression for neural networks with application to age estimation
• Silhouettes: A graphical aid to the interpretation and validation of cluster analysis
• A Coefficient of Agreement for Nominal Scales
• Analysis of a complex of statistical variables into principal components.

Discussions

• Excited to share my first article: “Machine learning for smell: Ordinal odor strength prediction of molecular perfumery components” on arXiv👃💻 🔗 arxiv.org/abs/2512.08683 Thanks from the support of [bsky, 2 points, 0 comments]

Related

• Navigating the Fragrance space Via Graph Generative Models And Predicting Odors
• Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules
• Molecule Generation and Optimization for Efficient Fragrance Creation
• Predicting human olfactory perception from chemical features of odor molecules
• Olfactory Label Prediction on Aroma-Chemical Pairs