Article
Medicine, Legal
Rito Chophi, Sweety Sharma, Jaskirandeep Kaur Jossan, Rajinder Singh
Summary: Cosmetic evidence can be utilized in crime investigations as associative evidence in court cases, providing a link between the suspect, victim, and crime scene. A study on eye-cosmetics analysis using ATR-FTIR spectroscopy shows differences in performance on various substrates for eyeliner and eyeshadow.
FORENSIC SCIENCE INTERNATIONAL
(2021)
Article
Automation & Control Systems
Xiaohong Chen, Xu Yang, Jing-wei Zhang, Hao Tang, Qing-hua Zhang, Ya-chen Wang, Zi-feng Jiang, Yan-ling Liu
Summary: Ink classification is the task of distinguishing unknown inks into different groups, while ink source prediction is the ability to predict the manufacturer or model of an unknown ink. This study presents an approach to predict the source of black inks using real-time mass spectrometry analysis and assesses the strength of the prediction using likelihood ratios. The results demonstrate the effectiveness of the proposed method in predicting ink source and calibrating the likelihood ratio.
JOURNAL OF CHEMOMETRICS
(2023)
Article
Chemistry, Analytical
Puneet Mishra, Ernst J. Woltering
Summary: This study improves the prediction of amylose content in grounded rice samples using four different wavenumber selection techniques, with VCPA showing the best performance and reducing prediction error by 19% with only 11 wavenumbers. The selected wavenumbers can aid in the development of low-cost multi-spectral sensors for amylose prediction in rice samples.
Article
Statistics & Probability
Daniel R. Kowal
Summary: This study emphasizes the importance of prediction for decision-making under uncertainty and the use of targeted prediction to optimize predictions for specific decision tasks. By designing a class of parameterized actions for Bayesian decision analysis, the study produces optimal, scalable, and simple targeted predictions. Results show that through careful use of the posterior predictive distribution, a procedure can be introduced to identify near-optimal targeted predictors, providing unique insights for accurate targeted prediction.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
History & Philosophy Of Science
Kino Zhao
Summary: The social sciences are facing the issue of non-representative samples, primarily comprised of undergraduate students from Euro-American institutions. Traditional design-based approach favors random sampling as the gold standard but encounters challenges in application. Advocating for the model-based approach, which focuses on balanced representative samples in composition regardless of how they were drawn, may be more suitable in social sciences for systematic evaluation and methodical improvement in resource-limited scientific environments.
Article
Physics, Multidisciplinary
Jakob Robnik, Uros Seljak
Summary: In hypothesis testing applications, mixed priors are common, with informative priors for some parameters but not for others. Bayesian methodology is helpful for informative priors, while frequentist hypothesis testing is better for cases where the prior is not completely known. Combining both methodologies by using the Bayes factor as a test statistic in frequentist analysis is recommended when only partial prior information is available.
Article
Multidisciplinary Sciences
Ming Zhang, Olivia Y. Xiao, Johan Lim, Xinlei Wang
Summary: This study proposes a novel goodness-of-fit test for meta-analysis of rare events. Compared to previous methods, it improves the ability to detect model misfits and has the advantages of clear conception and interpretation, incorporation of all data, and well-controlled Type I error.
SCIENTIFIC REPORTS
(2023)
Article
Plant Sciences
Yi Xiao, Jen Sloan, Chris Hepworth, Colin P. Osborne, Andrew J. Fleming, Xingyuan Chen, Xin-Guang Zhu
Summary: The Farquhar-von Caemmerer-Berry (FvCB) model is widely used for modeling photosynthesis, with this study introducing a new tool that utilizes Bayesian statistics for fitting photosynthetic parameters. Testing on experimental and synthetic data showed the tool's reliability and effectiveness, offering a user-friendly, interactive Bayesian script for parameter estimation.
PLANT CELL AND ENVIRONMENT
(2021)
Article
Medicine, Legal
Anais Hermelin, Loic Fabien, Julia Fischer, Nikola Saric, Genevieve Massonnet, Celine Burnier
Summary: The study investigated vaginal matrix residues using DRIFTS-FTIR and py-GC/MS, identifying proteins and lipids in the samples but no silicone residues. This has promising implications for forensic evidence interpretation, with further research needed to validate models and assess limitations in casework conditions.
FORENSIC SCIENCE INTERNATIONAL
(2021)
Article
Medicine, Legal
S. Huhtala, A. Nordgaard, B. Ahrens, I. Alberink, T. Korpinsalo, M. Bovens
Summary: In recent years, there has been a significant increase in the number of samples sent to forensic laboratories and the complexity of the drug situation. This has resulted in a large amount of data from chemical measurements. Forensic chemists face challenges in handling and analyzing this data to answer questions and establish connections between samples. Quality assessment steps are necessary before reporting chemometric results, which require considerations of the strengths, weaknesses, opportunities, and threats of chemometric methods. While powerful, chemometric methods have limitations and must be used in conjunction with other assessments.
FORENSIC SCIENCE INTERNATIONAL
(2023)
Article
Chemistry, Analytical
Danny Luarte, Ashwin Kumar Myakalwar, Marizu Velasquez, Jonnathan Alvarez, Claudio Sandoval, Rodrigo Fuentes, Jorge Yanez, Daniel Sbarbaro
Summary: This work presents a systematic methodology based on the Akaike information criterion (AIC) for selecting the wavelengths of LIBS spectra as well as the ANN model complexity, by combining prior knowledge and variable selection algorithms. Several variable selection algorithms are compared within the proposed methodology, showing that the proposed methodology is very effective for selecting wavelengths and model complexity in quantitative analyses based on ANNs and LIBS.
ANALYTICAL METHODS
(2021)
Article
Computer Science, Artificial Intelligence
Amy M. Crawford, Nicholas S. Berry, Alicia L. Carriquiry
Summary: This study focuses on characterizing and comparing handwritten documents by decomposing them into small graphical structures, measuring distance between these structures, and clustering them based on structural attributes. A Bayesian hierarchical model is then used to capture the tendency of a writer in producing certain graphs assigned to specific clusters.
STATISTICAL ANALYSIS AND DATA MINING
(2021)
Article
Computer Science, Artificial Intelligence
Xiaomin Zhang, Xiaojin Zhu, Po-Ling Loh
Summary: In this study, problems in penalized M-estimation were investigated with a focus on application in machine learning debugging. Algorithms for identifying buggy points and tuning parameter selection were proposed and rigorously analyzed. Moreover, a two-person game between a bug generator and a debugger was considered, leading to the development of a debugging strategy using Mixed Integer Linear Programming (MILP). Empirical results were provided to validate the theoretical findings and utility of the MILP strategy.
Article
Chemistry, Analytical
Amilton Moreira de Oliveira, Carlos Alberto Teixeira, Leandro Wang Hantao
Summary: In this study, comprehensive two-dimensional gas chromatography (GC x GC) combined with chemometrics was used for qualitative analysis of petroleum. The best performing GC x GC configuration was selected based on the elution pattern of hydrocarbons and the duty cycle of the method. By optimizing the forward and reverse fill/flush configurations, the best conditions for separating high-boiling point constituents were determined.
MICROCHEMICAL JOURNAL
(2022)
Article
Computer Science, Theory & Methods
Hakan Erdogmus
Summary: This tutorial provides a step-by-step illustration of Bayesian hypothesis testing in software engineering research, comparing it with the frequentist hypothesis testing approach. It demonstrates how Bayesian analysis can incrementally build evidence through a series of experiments and discusses the chief advantages and disadvantages in an applied manner.
ACM COMPUTING SURVEYS
(2023)
Article
Chemistry, Multidisciplinary
Ilya Kuselman, Francesca R. Pennecchi, Ricardo J. N. B. da Silva, David Brynn Hibbert
Summary: This study applies the Bayesian approach to define the risks of false decisions on the conformity of a multicomponent material or object due to measurement uncertainty. It explores the relationship between total risk and specific risk, and formulates a model for the total probability of false decisions. The paper also provides examples of risk evaluation for conformity assessment of various materials.
PURE AND APPLIED CHEMISTRY
(2021)
Article
Chemistry, Multidisciplinary
Zhifang Chai, Amares Chatt, Peter Bode, Jan Kucera, Robert Greenberg, David B. Hibbert
Summary: These recommendations provide a vocabulary of basic radioanalytical terms relevant to radioanalysis, nuclear analysis, and related techniques. They aim to facilitate understanding of compositional and structural analyses of materials in the context of nuclear processes, techniques, and effects.
PURE AND APPLIED CHEMISTRY
(2021)
Article
Chemistry, Multidisciplinary
Adriaan M. H. van der Veen, Juris Meija, Antonio Possolo, David Brynn Hibbert
Summary: This report provides guidelines for using standard atomic weights and discusses methods for calculating uncertainties of relative molecular masses and material-specific atomic weights.
PURE AND APPLIED CHEMISTRY
(2021)
Article
Chemistry, Multidisciplinary
David Brynn Hibbert, Ernst-Heiner Korte, Ulf Ornemark
Summary: Recommendations are made for standardizing metrological terminology in analytical chemistry, including defining concepts related to laboratory practice and quality assurance using references from extensive quality literature, particularly ISO standards.
PURE AND APPLIED CHEMISTRY
(2021)
Article
Chemistry, Multidisciplinary
Resat Apak, Antony Calokerinos, Shela Gorinstein, Marcela Alves Segundo, David Brynn Hibbert, Ilhami Gulcin, Sema Demirci Cekic, Kubilay Guclu, Mustafa Ozyurek, Saliha Esin Celik, Luis M. Magalhaes, Patricia Arancibia-Avila
Summary: This project aims to investigate the quenching chemistry of biologically important reactive oxygen and nitrogen species, evaluate the scavenging activity of antioxidants with existing methods, and emphasize the need for developing more refined assays to provide a comprehensive study of antioxidants and reactive species.
PURE AND APPLIED CHEMISTRY
(2022)
Article
Chemistry, Physical
Muhammad I. Ahmed, Lakshitha J. Arachchige, Zhen Su, David B. Hibbert, Chenghua Sun, Chuan Zhao
Summary: The electrochemical nitrogen reduction reaction (NRR) can be an alternative to the Haber-Bosch process for ammonia production, and single-atom catalysts have been proven effective in improving the reaction. By modulating the electronic structure of iron and tethering it to sulfur, the NRR performance can be enhanced.
Article
Chemistry, Analytical
Ricardo J. N. Bettencourt da Silva, Felipe Lourenco, D. Brynn Hibbert
Summary: The article presents a tool for setting multivariate acceptance limits applicable to correlated measurements. The tool allows for the decision about conformity of an item based on the simple comparison of the measured values with the acceptance limits. It is successfully applied to various conformity problems.
ANALYTICAL LETTERS
(2022)
Article
Environmental Sciences
Tamar Gadrich, Ilya Kuselman, Francesca R. Pennecchi, D. Brynn Hibbert, Anastasia A. Semenova, Pui Sze Cheow, Vladimir N. Naidenko
Summary: This study presents a case study on ordinal data from human organoleptic examination of drinking water obtained from 49 ecological laboratories. The application of the recently developed ORDANOVA method reveals its usefulness and reliability in analyzing and understanding categorical data related to the intensity of water odor and taste.
JOURNAL OF WATER AND HEALTH
(2022)
Article
Chemistry, Multidisciplinary
Igor M. Villa, Norman E. Holden, Antonio Possolo, Ryan Ben Ickert, David Brynn Hibbert, Paul R. Renne, Mauro L. Bonardi, Paul De Bievre
Summary: The IUPAC-IUGS joint Task Group has evaluated the half-lives of six long-lived radioactive nuclides and recommends three robust estimates. However, for other nuclides, further investigations are needed to resolve uncertainties.
PURE AND APPLIED CHEMISTRY
(2022)
Article
Food Science & Technology
Tamar Gadrich, Francesca R. R. Pennecchi, Ilya Kuselman, D. Brynn Hibbert, Anastasia A. A. Semenova, Pui Sze Cheow
Summary: The newly developed statistical technique of two-way ordinal analysis of variation (ORDANOVA) was applied for the first time to sensory responses in combination with multinomial ordered logistic regression of a response category vs. chemical composition. A case study on sausage samples showed statistically significant differences between producers' samples and insignificant differences between experts' responses related to the same sample. The influence of chemical composition on sensory responses was modeled using multinomial ordered logistic regression, providing insights into understanding food quality properties.
JOURNAL OF FOOD QUALITY
(2022)
Review
Chemistry, Multidisciplinary
Jan Kaiser, David Brynn Hibbert, Juergen Stohner
PURE AND APPLIED CHEMISTRY
(2023)
Article
Chemistry, Analytical
Danielle Bennett, Xueqian Chen, Gregory J. Walker, Sacha Stelzer-Braid, William D. Rawlinson, D. Brynn Hibbert, Richard D. Tilley, J. Justin Gooding
Summary: Plasmonic nanoparticles in dimer format are used for single molecule sensing, where the interaction with hairpin DNA leads to a shift in localized surface plasmon resonance. Spectroscopy may detect this shift, but point-of-care devices require a faster analysis method. By using dark-field imaging and digital analysis, the plasmonic resonance shift of thousands of dimer structures can be measured in minutes. The challenge is separating dimers from non-specifically bound clusters to achieve accurate results. The LAB-based classifier algorithm demonstrated the highest accuracy for this digital separation.
ANALYTICAL CHEMISTRY
(2023)
Correction
Chemistry, Analytical
Danielle Bennett, Xueqian Chen, Gregory J. Walker, Milad Mehdipour, Sacha Stelzer-Braid, William D. Rawlinson, D. Brynn Hibbert, Richard D. Tilley, J. Justin Gooding
ANALYTICAL CHEMISTRY
(2023)
Article
Chemistry, Multidisciplinary
David G. Shaw, Ian Bruno, Stuart Chalk, Glenn Hefter, David Brynn Hibbert, Robin A. Hutchinson, M. Clara F. Magalhaes, Joseph Magee, Leah R. McEwen, John Rumble, Gregory T. Russell, Earle Waghorne, Thomas Walczyk, Timothy J. Wallington
Summary: The International Union of Pure and Applied Chemistry (IUPAC) has a strong commitment to supporting the compilation and evaluation of chemical data through various means. The establishment of the IUPAC Interdivisional Subcommittee on Critical Evaluation of Data aims to provide guidance in this area. This first report focuses on defining general principles and best practices for data evaluation in chemistry.
PURE AND APPLIED CHEMISTRY
(2023)
Article
Chemistry, Multidisciplinary
Francesca R. Pennecchi, Ilya Kuselman, D. Brynn Hibbert
Summary: This study adapts a Bayesian multivariate approach to evaluate the risks of false decisions on conformity of chemical composition under measurement uncertainty, taking into account a mass balance constraint. By considering correlations, a technique for appropriate risk evaluation is discussed, and Monte Carlo method is applied for calculations.
PURE AND APPLIED CHEMISTRY
(2023)
Article
Automation & Control Systems
Haifei Peng, Jian Long, Cheng Huang, Shibo Wei, Zhencheng Ye
Summary: This paper proposes a novel multi-modal hybrid modeling strategy (GMVAE-STA) that can effectively extract deep multi-modal representations and complex spatial and temporal relationships, and applies it to industrial process prediction.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2024)