Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Recolector de Cienci...arrow_drop_down
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao
image/svg+xml Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao Closed Access logo, derived from PLoS Open Access logo. This version with transparent background. http://commons.wikimedia.org/wiki/File:Closed_Access_logo_transparent.svg Jakob Voss, based on art designer at PLoS, modified by Wikipedia users Nina and Beao
versions View all 2 versions
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

Prediction of substances’ chemical properties based on their composition

Authors: Arribas Basora, Laura;

Prediction of substances’ chemical properties based on their composition

Abstract

Esta tesis se desarrolló en un entorno industrial, trabajando como Data Scientist en Basetis, una consultora tecnológica con sede en Barcelona. La tesis detalla los experimentos y resultados obtenidos por el equipo de IA de Basetis en colaboración con Lucta —una empresa global especializada en la síntesis de fragancias, sabores y aditivos para piensos. En particular, el objetivo de este trabajo es predecir con alta precisión las principales propiedades químicas —densidad, índice de refracción y punto de inflamación— de una serie de artículos utilizados por Lucta. Se entrenan varios modelos predictivos, que incluyen tanto modelos analíticos basados en física como algoritmos de aprendizaje automático para datos tabulares de propósito general, como TabPFN y XGBoost. Con estos modelos, se logró una alta precisión en la tarea de predicción.

Aquesta tesi es va desenvolupar en un entorn industrial, treballant com a Data Scientist a Basetis, una consultora tecnològica amb seu a Barcelona. La tesi descriu els experiments i els resultats obtinguts per l'equip d’IA de Basetis en col·laboració amb Lucta —una empresa global especialitzada en la síntesi de fragàncies, sabors i additius per a pinsos. En concret, l’objectiu d’aquest treball és predir amb alta precisió les principals propietats químiques —densitat, índex de refracció i punt d’ignició— d’una sèrie d’articles utilitzats per Lucta. S’entrenen diversos models predictius, que inclouen tant models analítics basats en física com algoritmes d’aprenentatge automàtic per a dades tabulars d’ús general, com TabPFN i XGBoost. Mitjançant aquests models, s’ha aconseguit una alta precisió en la tasca de predicció.

This thesis was developed in industry, working as a Data Scientist for Basetis, a Barcelona-based technological consultancy firm. The thesis details the experiments and results obtained by the Basetis AI team in collaboration with Lucta—a global company specializing in the synthesis of fragrances, flavors, and feed additives. In particular, the objective of this work is to predict with high accuracy the key chemical properties -- density, refractive index and flashpoint -- of a series of articles used by Lucta. We train a series of predictive models, which include both analytic physics-based models and general purpose tabular Machine Learning algorithms, like TabPFN and XGBoost. Using these, we accomplished high accuracy in the prediction task.

Related Organizations
Keywords

Machine Learning, Àrees temàtiques de la UPC::Matemàtiques i estadística, Classificació AMS::68 Computer science::68T Artificial intelligence, Machine learning, Aprenentatge automàtic, Predictive analytics, Modelització predictiva (Estadística), Classificació AMS::90 Operations research, mathematical programming::90C Mathematical programming, XGBoost, data augmentation

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
Upload OA version
Are you the author of this publication? Upload your Open Access version to Zenodo!
It’s fast and easy, just two clicks!