非破坏性光谱传感器的多变量校准，特别关注食品应用：验证问题和指南

IF 12 1区化学 Q1 CHEMISTRY, ANALYTICAL

Trends in Analytical Chemistry Pub Date : 2025-08-07 DOI:10.1016/j.trac.2025.118410

Tom Fearn , Claudia Beleites , Juan Antonio Fernández Pierna , Vincent Baeten , Martin Lagerholm , Jean-Michel Roger , Anastasios Koidis

{"title":"非破坏性光谱传感器的多变量校准，特别关注食品应用：验证问题和指南","authors":"Tom Fearn , Claudia Beleites , Juan Antonio Fernández Pierna , Vincent Baeten , Martin Lagerholm , Jean-Michel Roger , Anastasios Koidis","doi":"10.1016/j.trac.2025.118410","DOIUrl":null,"url":null,"abstract":"<div><div>Multivariate calibration methods have enabled the use of non-destructive spectral sensors in a wide range of applications but carry a risk of overfitting to the available training samples. For this reason, the prediction of unseen samples plays a vital role both in tuning the prediction algorithm and in assessing its performance, two activities that need to be carefully distinguished. Methods employed include data-splitting, cross-validation, and the use of genuinely independent sets of data. These approaches are described and some common issues with them are identified. The focus is on food applications but the methods discussed are widely used in other areas.</div></div>","PeriodicalId":439,"journal":{"name":"Trends in Analytical Chemistry","volume":"192 ","pages":"Article 118410"},"PeriodicalIF":12.0000,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Multivariate calibration of non-destructive spectral sensors with a particular focus on food applications: Validation issues and guidelines\",\"authors\":\"Tom Fearn , Claudia Beleites , Juan Antonio Fernández Pierna , Vincent Baeten , Martin Lagerholm , Jean-Michel Roger , Anastasios Koidis\",\"doi\":\"10.1016/j.trac.2025.118410\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Multivariate calibration methods have enabled the use of non-destructive spectral sensors in a wide range of applications but carry a risk of overfitting to the available training samples. For this reason, the prediction of unseen samples plays a vital role both in tuning the prediction algorithm and in assessing its performance, two activities that need to be carefully distinguished. Methods employed include data-splitting, cross-validation, and the use of genuinely independent sets of data. These approaches are described and some common issues with them are identified. The focus is on food applications but the methods discussed are widely used in other areas.</div></div>\",\"PeriodicalId\":439,\"journal\":{\"name\":\"Trends in Analytical Chemistry\",\"volume\":\"192 \",\"pages\":\"Article 118410\"},\"PeriodicalIF\":12.0000,\"publicationDate\":\"2025-08-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Trends in Analytical Chemistry\",\"FirstCategoryId\":\"1\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S016599362500278X\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, ANALYTICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Trends in Analytical Chemistry","FirstCategoryId":"1","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S016599362500278X","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, ANALYTICAL","Score":null,"Total":0}

引用次数: 0

摘要

多元校准方法使无损光谱传感器在广泛应用中得以使用，但存在与可用训练样本过拟合的风险。由于这个原因，对未见样本的预测在调整预测算法和评估其性能方面都起着至关重要的作用，这两个活动需要仔细区分。采用的方法包括数据分割、交叉验证和使用真正独立的数据集。本文描述了这些方法，并指出了它们的一些常见问题。重点是食品应用，但所讨论的方法已广泛应用于其他领域。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Multivariate calibration of non-destructive spectral sensors with a particular focus on food applications: Validation issues and guidelines

Multivariate calibration methods have enabled the use of non-destructive spectral sensors in a wide range of applications but carry a risk of overfitting to the available training samples. For this reason, the prediction of unseen samples plays a vital role both in tuning the prediction algorithm and in assessing its performance, two activities that need to be carefully distinguished. Methods employed include data-splitting, cross-validation, and the use of genuinely independent sets of data. These approaches are described and some common issues with them are identified. The focus is on food applications but the methods discussed are widely used in other areas.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Trends in Analytical Chemistry 化学-分析化学

CiteScore

20.00

自引率

4.60%

发文量

257

审稿时长

3.4 months

期刊介绍： TrAC publishes succinct and critical overviews of recent advancements in analytical chemistry, designed to assist analytical chemists and other users of analytical techniques. These reviews offer excellent, up-to-date, and timely coverage of various topics within analytical chemistry. Encompassing areas such as analytical instrumentation, biomedical analysis, biomolecular analysis, biosensors, chemical analysis, chemometrics, clinical chemistry, drug discovery, environmental analysis and monitoring, food analysis, forensic science, laboratory automation, materials science, metabolomics, pesticide-residue analysis, pharmaceutical analysis, proteomics, surface science, and water analysis and monitoring, these critical reviews provide comprehensive insights for practitioners in the field.