Model Selection Research Articles

Machine-Learning Prediction of Curie Temperature from Chemical Compositions of Ferromagnetic Materials.

Room-temperature ferromagnets are high-value targets for discovery given the ease by which they could be embedded within magnetic devices. However, the multitude of potential interactions among magnetic ions and their surrounding environments renders the prediction of thermally stable magnetic properties challenging. Therefore, it is vital to explore methods that can effectively screen potential candidates to expedite the discovery of novel ferromagnetic materials within highly intricate feature spaces. To this end, we explore machine-learning (ML) methods as a means to predict the Curie temperature (Tc) of ferromagnetic materials by discerning patterns within materials databases. This study emphasizes the importance of feature analysis and selection in ML modeling and demonstrates the efficacy of our gradient-boosted statistical feature-selection workflow for training predictive models. The models are fine-tuned through Bayesian optimization, using features derived solely from the chemical compositions of the materials data, before the model predictions are evaluated against literature values. We have collated ca. 35,000 Tc values and the performance of our workflow is benchmarked against state-of-the-art algorithms, the results of which demonstrate that our methodology is superior to the majority of alternative methods. In a 10-fold cross-validation, our regression model realized an R2 of (0.92 ± 0.01), an MAE of (40.8 ± 1.9) K, and an RMSE of (80.0 ± 5.0) K. We demonstrate the utility of our ML model through case studies that forecast Tc values for rare-earth intermetallic compounds and generate magnetic phase diagrams for various chemical systems. These case studies highlight the importance of a systematic approach to feature analysis and selection in enhancing both the predictive capability and interpretability of ML models, while being devoid of potential human bias. They demonstrate the advantages of such an approach over a mere reliance on algorithmic complexity and a black-box treatment in ML-based modeling within the domain of computational materials science.

Quantum Mechanical Cluster Models for Calculations on Enzymatic Reaction Mechanisms: Set-Up and Accuracy.

Enzymes turnover substrates into products with amazing efficiency and selectivity and as such have great potential for use in biotechnology and pharmaceutical applications. However, details of their catalytic cycles and the origins surrounding the regio- and chemoselectivity of enzymatic reaction processes remain unknown, which makes the engineering of enzymes and their use in biotechnology challenging. Computational modelling can assist experimental work in the field and establish the factors that influence the reaction rates and the product distributions. A popular approach in modelling is the use of quantum mechanical cluster models of enzymes that take the first- and second coordination sphere of the enzyme active site into consideration. These QM cluster models are widely applied but often the results are dependent on model choice and selection. Herein, we show that QM cluster models can produce highly accurate results that reproduce experimental product distributions and free energies of activation, regarded that large cluster models with >300 atoms are used. In this tutorial review, we give general guidelines on the set-up and applications of the QM cluster method and discuss its accuracy and reproducibility. Finally, several representative QM cluster model examples on metal-containing enzymes are presented, which highlight the strength of the approach.

Artificial Intelligence Applications in English News Translation: Strategies and Research

The rapid advancement of Artificial Intelligence (AI) is significantly transforming the field of translation, especially in English news translation. This paper aims to explore the application strategies of AI in English news translation and its current research status. First, it introduces the basic concepts of AI and machine translation, reviews the development history of machine translation technology, and analyzes the unique requirements of news translation. Then, it discusses key application strategies of AI in news translation, including data collection and processing, model selection and training, translation quality assessment, and real-time translation technology. Additionally, the paper analyzes major challenges in AI applications, such as translation accuracy, cultural and contextual understanding, and data privacy and security, and proposes corresponding solutions. Finally, it explores future development trends in AI translation technology, including technological advancements, application expansion, and ethical and societal impacts. This paper aims to provide comprehensive references and in-depth insights for researchers in related fields.

62 Prognostic biomarker development using deep learning and spatial proteomics in papillary renal cell carcinoma

Abstract Background Papillary renal cell carcinoma (pRCC) is a prevalent but understudied kidney cancer pathology, constituting 15-20% of cases. While many patients achieve curative therapy using nephrectomy, others go on to develop progressive, metastatic disease with poor overall prognosis. In order to improve the current standards of care for pRCC patients, it is critical to identify new methods to identify high-risk patients likely to require more intense and risk-modulated follow-up. Our goals in this project are two-fold: (1) to identify prognostic patient clusters based on spatial architecture of tumor tissue, and (2) to identify a minimal subset of markers to accurately predict the defined clusters for clinical translation. Methods A tissue microarray of 100 pRCC patients was assayed using multiplex immunofluorescence with a 31-antibody panel with immune- and cancer-related proteins. We have developed novel algorithms to identify spatial neighborhoods and features. These features are then used to identify unsupervised clusters of patients. This new approach to classifying patients allows for a more specific and comprehensive description of each patient tumor’s spatial composition and association with diverse clinical outcomes. Next, we developed attention-based deep learning models capable of predicting these spatially-defined clusters directly from the underlying immunofluorescence images and subsequently profiled these for interpretability by retraining based on individual channels. We assessed each model iteratively on a validation set and a held-out test set for final model selection. We focussed on developing an accurate model with the minimal number of channels to enable development of a cost-efficient biomarker assay for the clinic. Results We performed a systematic analysis of the cellular and spatial phenotypes for each patient in our cohort and identified six (6) major spatial clusters. These clusters are based on the spatial neighborhood composition of the tumors (Fig 1A) and describe the composite effects of cell-cell interactions within a patient tumor. We identified one cluster in particular (cluster #6, mainly associated with spatial interaction of CD163+ macrophages) with a significantly worse prognosis than the other clusters. Using our advanced deep learning models, we have been able to achieve excellent validation accuracy in predicting a variety of clinical phenotypes and spatial features, including patient clusters and histological grade. We wished to then reduce the set of required features and began our efforts in predicting tumor vs. normal based on individual immunofluorescence channels and identified 6 channels with >75% individual predictive accuracy (Fig. 1B). Compositing all 6 markers further improves our accuracy to 90%. This success paves the way for exciting future prospects as we work to extend these efforts to identify a minimal clinically-applicable subset of protein markers. Conclusions The combination of computational spatial proteomics and deep learning models has the potential to identify new biomarkers specific to the spatial organization of patient tumors. We have identified a particular patient cluster with multiple associations with CD163+ macrophages demonstrating poor prognosis; selecting these patients may allow for tailored and more informed patient care. We are also working to enhance our deep learning models to circumvent the need for stepwise spatial analysis and, instead, allow for direct-from-image assessment of patients for particular spatial features indicative of various prognostic indicators. Our identified set of 6 markers serves as a basis for further refinements as we work to identify a minimal set of relevant markers able to predict various clinical outcomes from immunofluorescence imaging, offering a promising new avenue in the field of cancer pathology and biomarker development. DOD CDMRP Funding: yes

Estimating highest capacity propulsion performance using backward-directed force during walking evaluation for individuals with acquired brain injury

There are over 5.3 million Americans who face acquired brain injury (ABI)-related disability as well as almost 800,000 who suffer from stroke each year. To improve mobility and quality of life, rehabilitation professionals often focus on walking recovery soon after hospital discharge for ABI. Reduced propulsion capacity (force output of the lower limbs to counteract ground reaction forces) negatively impacts walking ability and complicates recovery during rehabilitation for brain injured people. We describe a method, using backward-directed resistance (BDR) in a robotic-based treadmill device, to allow measurement of maximum walking propulsion force (MWPF) that is not otherwise possible during overground walking assessment. Our objective was to test the construct validity of a maximum walking propulsion force (MWPF) measure that reflects a person’s propulsive strength against applied BDR, while walking on a robotic treadmill-based device for participants with acquired brain injury (ABI). Our study enrolled 14 participants with ABI at an in inpatient rehabilitation in Galveston, TX from 8/1/21 − 4/31/22. The range of weight-adjusted MWPF was 2.6–27.1% body weight (%BW), mean 16.5 ± 8.4%BW, reflecting a wide range of propulsive force capability. The strongest correlation with overground tests was between the 6-minute walk test (6-MWT) distance and the MWPF values (r = 0.83, p < 0.001) with moderate correlations between the 10-meter walk tests at comfortable (CWS) and fast speeds (FWS). The Five Times Sit-to-Stand (used as a standard clinical measure of functional lower extremity strength) and MWPF tests were poorly correlated (r = 0.26, p = 0.4). Forward model selection included 6-MWT distance, age, and overground CWS as significant partial predictors of MWPF. We conclude that this novel MWPF measure is a valid representation of maximum propulsive force effort during walking for people post-ABI. Additional research could help determine the impact of interventions designed to increase propulsive force generation during rehabilitation training to improve overground walking performance.

Spatiotemporal patterns and association with climate for malaria elimination in Lao PDR: a hierarchical modelling analysis with two-step Bayesian model selection

BackgroundThe government of Lao PDR has increased efforts to control malaria transmission in order to reach its national elimination goal by 2030. Weather can influence malaria transmission dynamics and should be considered when assessing the impact of elimination interventions but this relationship has not been well characterized in Lao PDR. This study examined the space–time association between climate variables and Plasmodium falciparum and Plasmodium vivax malaria incidence from 2010 to 2022.MethodsSpatiotemporal Bayesian modelling was used to investigate the monthly relationship, and model selection criteria were used to evaluate the performance of the models and weather variable specifications. As the malaria control and elimination situation was spatially and temporally dynamic during the study period, the association was examined annually at the provincial level.ResultsMalaria incidence decreased from 2010 to 2022 and was concentrated in the southern regions for both P. falciparum and P. vivax. Rainfall and maximum humidity were identified as most strongly associated with malaria during the study period. Rainfall was associated with P. falciparum incidence in the north and central regions during 2010–2011, and with P. vivax incidence in the north and central regions during 2012–2015. Maximum humidity was persistently associated with P. falciparum and P. vivax incidence in the south.ConclusionsMalaria remains prevalent in Lao PDR, particularly in the south, and the relationship with weather varies between regions but was strongest for rainfall and maximum humidity for both species. During peak periods with suitable weather conditions, vector control activities and raising public health awareness on the proper usage of intervention measures, such as indoor residual spraying and personal protection, should be prioritized.

Towards Automated Model Selection for Wind Speed and Solar Irradiance Forecasting

Given the recent increase in demand for electricity, it is necessary for renewable energy sources (RESs) to be widely integrated into power networks, with the two most commonly adopted alternatives being solar and wind power. Nonetheless, there is a significant amount of variation in wind speed and solar irradiance, on both a seasonal and a daily basis, an issue that, in turn, causes a large degree of variation in the amount of solar and wind energy produced. Therefore, RES technology integration into electricity networks is challenging. Accurate forecasting of solar irradiance and wind speed is crucial for the efficient operation of renewable energy power plants, guaranteeing the electricity supply at the most competitive price and preserving the dependability and security of electrical networks. In this research, a variety of different models were evaluated to predict medium-term (24 h ahead) wind speed and solar irradiance based on real-time measurement data relevant to the island of Crete, Greece. Illustrating several preprocessing steps and exploring a collection of “classical” and deep learning algorithms, this analysis highlights their conceptual design and rationale as time series predictors. Concluding the analysis, it discusses the importance of the “features” (intended as “time steps”), showing how it is possible to pinpoint the specific time of the day that most influences the forecast. Aside from producing the most accurate model for the case under examination, the necessity of performing extensive model searches in similar studies is highlighted by the current work.

ROBUST GEOGRAPHICALLY WEIGHTED REGRESSION WITH LEAST ABSOLUTE DEVIATION (LAD) ESTIMATION AND M-ESTIMATION ON GRDP OF WEST JAVA PROVINCE

Geographically Weighted Regression (GWR) is an analytical method for data that contains spatial heterogeneity effects. However, parameter estimation in the GWR model has a weakness, namely it is prone to outliers and can cause the parameter estimation to be biased. This can be overcome by the Robust Geographically Weighted Regression (RGWR) method which is more robust against the presence of outliers. This method is suitable for Gross Regional Domestic Product (GRDP) data in West Java Province, which contains outliers and also has spatial effects. The data used in this study are secondary data obtained from the Central Statistics Agency (BPS) of West Java Province. The purpose of this study is to compare the Robust Geographically Weighted Regression (RGWR) method with the Least Absolute Deviation (LAD) Estimation and M-estimation and also to find out the factors that affect the Gross Regional Domestic Product (GRDP) in West Java Province in 2021 based on the model resulting from. Selection of the best model is seen based on the value of the coefficient of determination (R2) and Mean Squared of Error (MSE). The research results show that the Robust Geographically Weighted Regression (RGWR) method with M-estimation is much more effective in estimating the distribution of GRDP in West Java Province in 2021, seen from the larger coefficient of determination and the smaller Mean Square Error (MSE). The variables that have a significant influence on GRDP in West Java Province in 2021 are the variables of foreign investment and local income.

Predicting seismic sustainability for a complex CHESST interaction by AHP using LWST

Underground and over-ground structures are affected by the external forces (tsunami, earthquake, storm, etc.) that are directly related to the environment and change in climate. For this reason, a complex CHESST (Climate-Human-Environment-Structure-Soil-Tunnel) interaction model is proposed in this research which is categorized based on the structure types (building, bridge, elevated, road/rail). The best CHESST interaction model is selected based on the AHP (Analytic Hierarchy Process) by using LWST (Logical Weightage Selection Technique). The LWST depends on the proposed modified logical scale. The logical weightage matrices are prepared for the three criteria (Environment and Climate, Social, and Economic), three sub-criteria under each criteria, and goal (selection of best CHESST interaction model). The weightage selection technique is validated based on the acceptable rate of decision change sensitivity. The elevated (alternative of the CHESST interaction model) shows the maximum value of seismic sustainability index (SSI) among all alternatives based on the AHP results. Also, the maximum fluctuation rate of strain is found for the elevated structure of the CHESST interaction model from the shake table test results. The most seismic sustainable CHESST interaction model is elevated expressway according to the SSI because of showing highest priority of 0.368. However, there is a scope to enhance this research in future by considering more alternatives, criteria, sub-criteria, etc.

Robust estimation of global horizontal irradiance with modified fuzzy regression functions with a noise cluster in Australia

The utilization of solar energy is picking up speed to counter climate change. New large-scale photovoltaic power stations are being constructed to increase solar utilization in the energy mix. A critical input of site selection for solar farms is the solar energy generation potential at a given location. Various physical and satellite-based inversion models are proposed to estimate the solar irradiation reaching the ground at potential locations, based on the meteorological features. However, the meteorological features generally contain outlier observations that distract the solar irradiation estimation models. To address this challenge, this study employs a robust fuzzy regression functions framework against the outliers to estimate the global horizontal irradiance (GHI) in Australia. Our framework is benchmarked with support vector machines, deep neural networks, and an adaptive network-based fuzzy inference system, and better GHI estimation performance is observed. The proposed framework provides 24 %, 18 %, and 23 % gain over the second-best method in terms of the rescaled mean absolute error, absolute percentage bias and rescaled root-mean-squared error. Monthly and annual GHI maps are created for Australia and compared to those from NASA POWER GHI estimates and Solargis annual GHI estimates. Our framework has an error range between 0.075 % and 2.9 % when validated against ground measurements. It provides at least an average of 40% lower monthly and annual error rates than POWER. This rate of gain rises to 69% when compared to Solargis. Our maps are not impacted by terrestrial characteristics and clear-sky conditions. This study’s results are beneficial in site selection and construction of high-precision GHI estimation models for practitioners.

Private cloud deployment model selection for cost efficiency: a business perspective

Private cloud deployment model selection for cost efficiency: a business perspective

Fostering the Use of Reflective Learning Strategies in Teaching Big Data Analytics and Data Science: Enhancing the Quality of Epidemiological Evidence and Decision-Making

PURPOSEThis study investigated the frequency of use and perceived effectiveness of reflective learning strategies among university educators and practitioners in big data analytics and data science, including epidemiologists. Reflective learning fosters critical thinking and problem-solving skills essential for navigating the complexities of big data. It encourages systematic evaluation of methodologies and the identification and mitigation of biases. These skills are particularly relevant in epidemiology, where the prediction and management of disease outbreaks, as well as public health decisions, increasingly depend on big data analytics. METHODOLOGYA cross-sectional design was employed with a diverse sample of 93 participants from multiple countries, averaging 15 years of teaching and practice experience. Informed by the Diffusion of Innovation Theory and the Theory of Planned Behavior, data were collected in Fall 2023 using an online questionnaire. The questionnaire focused on personal disposition, contextual favorability, and characteristics of reflective learning as predictors of the frequency and perceived effectiveness of reflective learning, analyzed using multiple regression modeling. Acceptable levels of reliability and criterion validity were achieved. This study was approved by the Ethics Panel of Canterbury Christ Church University, UK. RESULTSIn general, reflective learning strategies were perceived as effective for enhancing cognitive skills, such as critical thinking and problem-solving, but less so for technical skills like algorithm and model selection. Preferred methods included real-world applications, problem-based learning, and collaborative projects. Frequency of use was positively correlated with perceived effectiveness. Personal disposition and characteristics of reflective learning scores were significant predictors of frequency of use (R-squared = 0.32), with older participants being less engaged. The characteristics of reflective learning score was the sole significant predictor of perceived effectiveness, though with a modest predictive value (R-squared = 0.09). Contextual favorability did not significantly predict either frequency of use or perceived effectiveness of reflective learning strategies, implying a lesser role for external factors as drivers of engagement. CONCLUSIONSThis study underscores the importance of reflective learning in enhancing analytical and problem-solving skills, and suggests that intrinsic motivation may be a key factor underpinning the use of reflective learning strategies by educators and practitioners in big data analytics and data science. These findings have major implications for curriculum and professional development, particularly for epidemiologists who must continuously adapt to new data and methodologies to effectively address public health challenges.

A brief note on the random-effects meta-analysis model and its relationship to other models

Meta-analysis is a statistical method for combining quantitative results across studies. A fundamental decision in undertaking a meta-analysis is choosing an appropriate model for analysis. This is the second of two companion articles which have the joint aim of describing the different meta-analysis models. In the first article, we focused on the common-effect (also known as fixed-effect [singular]) model, and in this article, we focus on the random-effects model. We describe the key assumptions underlying the random-effects model, how it is related to the common-effect and fixed-effects [plural] models, and present some of the arguments for selecting one model over another. We outline some of the methods for fitting a random-effects model. Finally, we present an illustrative example to demonstrate how the results can differ depending on the chosen model and method. Understanding the assumptions of the different meta-analysis models, and the questions they address, is critical for meta-analysis model selection and interpretation.

Efficient predictive modeling of resilient modulus in stabilized clayey soil using automated machine learning

Pavement subgrade design relies on the resilient modulus (Mr) to analyze structural response to vehicle-like loading. Adding stabilizers to the soil subgrade makes estimating Mr difficult and resource-intensive. This study uses an automated machine learning (ML) strategy to predict the Mr of stabilized clayey soil using recycled plastic waste. The proposed method automates model selection and hyperparameter tuning, making it a feasible alternative to tedious ML modeling and costly laboratory testing. From extensive laboratory investigation involving 3285 experimental data points, the automated ML model using Bayesian optimization evaluates ensembles, support vector machine (SVM), neural network (NET), decision trees, (TREE), and Gaussian process (GP) regression models, identifying the best model based on cross-validation mean squared error (MSE). Bayesian optimization explores hyperparameter spaces to find optimal configurations, enhancing the accuracy, scalability, and reliability of the prediction model. The optimization process yielded the best results for the ensemble least square boost (LSBoost) model with a cross-validation mean squared error (MSE) value of 6.723×10−29. The optimized ML model’s performance is measured using R2 and adjusted R2. The LSboost model’s R2 and adjusted R2 values of 0.9999 suggested overfitting, prompting further investigations using performance metrics like root mean squared error (RMSE), mean absolute error (MSE), and probability density function (PDF) for normalized absolute error (NAE) for training and testing datasets for the predictive ML model. The small RMSE and MAE (0.0049 and 0.0005) values and symmetrical NAE distribution of the proposed ML model demonstrate its high accuracy and generalization capabilities. The proposed model was subsequently tested on the new, unseen data and achieved predictions with an error rate of 0.24 %. This confirms the proposed ML model’s superiority over conventional deformation models, making it ideal for reliable geotechnical engineering applications.

Total-Body Dynamic Imaging and Kinetic Modeling of [18F]F-AraG in Healthy Individuals and a Non-Small Cell Lung Cancer Patient Undergoing Anti-PD-1 Immunotherapy.

Immunotherapies, especially checkpoint inhibitors such as anti-programmed cell death protein 1 (anti-PD-1) antibodies, have transformed cancer treatment by enhancing the immune system's capability to target and kill cancer cells. However, predicting immunotherapy response remains challenging. 18F-arabinosyl guanine ([18F]F-AraG) is a molecular imaging tracer targeting activated T cells, which may facilitate therapy response assessment by noninvasive quantification of immune cell activity within the tumor microenvironment and elsewhere in the body. The aim of this study was to obtain preliminary data on total-body pharmacokinetics of [18F]F-AraG as a potential quantitative biomarker for immune response evaluation. Methods: The study consisted of 90-min total-body dynamic scans of 4 healthy subjects and 1 non-small cell lung cancer patient who was scanned before and after anti-PD-1 immunotherapy. Compartmental modeling with Akaike information criterion model selection was used to analyze tracer kinetics in various organs. Additionally, 7 subregions of the primary lung tumor and 4 mediastinal lymph nodes were analyzed. Practical identifiability analysis was performed to assess the reliability of kinetic parameter estimation. Correlations of the SUVmean, the tissue-to-blood SUV ratio (SUVR), and the Logan plot slope (K Logan) with the total volume of distribution (V T) were calculated to identify potential surrogates for kinetic modeling. Results: Strong correlations were observed between K Logan and SUVR with V T, suggesting that they can be used as promising surrogates for V T, especially in organs with a low blood-volume fraction. Moreover, practical identifiability analysis suggested that dynamic [18F]F-AraG PET scans could potentially be shortened to 60 min, while maintaining quantification accuracy for all organs of interest. The study suggests that although [18F]F-AraG SUV images can provide insights on immune cell distribution, kinetic modeling or graphical analysis methods may be required for accurate quantification of immune response after therapy. Although SUVmean showed variable changes in different subregions of the tumor after therapy, the SUVR, K Logan, and V T showed consistent increasing trends in all analyzed subregions of the tumor with high practical identifiability. Conclusion: Our findings highlight the promise of [18F]F-AraG dynamic imaging as a noninvasive biomarker for quantifying the immune response to immunotherapy in cancer patients. Promising total-body kinetic modeling results also suggest potentially wider applications of the tracer in investigating the role of T cells in the immunopathogenesis of diseases.

An ensemble approach for enhanced Day-Ahead price forecasting in electricity markets

Electricity price forecasting (EPF) is a crucial aspect of daily trading operations, enabling market participants to make informed decisions regarding their bidding strategies. This paper explores a day-ahead price forecasting system that harnesses the potential of multiple machine learning (ML) models and their synergistic integration. This approach is designed to capitalize on the strengths of these models while also accounting for the unique characteristics of energy markets. For this purpose, several aggregation models were developed combining the predictions from ML models based on historical evaluations of their performance. The main objective of this approach is to enhance prediction accuracy by shifting the focus away from rigid model selection and instead prioritizing a data-centric approach, by focusing on data quality rather than rigid model selection. As a case study, the German energy market was examined due to its pivotal role within the EU system. The experimental results from this study provide valuable insights into the proposed system’s effectiveness and functionality.

Inference on noncooperative entry deterrence

AbstractThis study empirically investigates strategic entry‐deterrence behavior under oligopolistic competition. I develop a structural econometric model describing incumbents' entry‐deterrence behavior based on the framework of Gilbert and Vives. I show theoretically that incumbents' marginal costs are interval‐identified under the assumption that incumbents deter entry in equilibrium. The structural model is estimated using data from the Japanese aluminum smelting industry. A Vuong‐type model selection test utilizing an instrument demonstrates that the entry‐deterrence model is more consistent with the data than an ordinary Cournot competition model without entry threats.

Machine learning methods in evaluating the impact of economic factors on the consumer price index in Albania

The Consumer Price Index (CPI) in Albania is a measure of inflation that tracks changes in the prices of a basket of goods and services typically purchased by urban households in the country. It is a vital economic indicator used to assess changes in the cost of living and the overall price level in Albania. There are several factors that affect the levels and progress of the CPI, among them we have chosen: Euro/Lek and USD/Lek exchange rates, import levels, the monetary base, and salary data, from January 2007 to September 2023. In this paper, we investigate the efficiency of machine learning methods in determining the factors that have the greatest impact on the CPI. In our analysis, we assess the effectiveness of decision-tree models, Random Forest and XGBoost algorithms, in predicting the CPI behavior in Albania. Based on our empirical findings, we conclude that the monetary base and wages play a crucial role in influencing the CPI, with imports and exchange rates following closely in significance. Additionally, our results indicate that the Random Forest model demonstrates superior accuracy and demands less parameter tuning time compared to the alternatives. This research underscores the critical role of model selection in achieving precision and dependability in CPI forecasting. It underscores the immense potential of machine learning models in enhancing forecasting accuracy. The implications of this study are significant, as they can foster the creation of more precise and dependable forecasting models, equipping policymakers with a deeper understanding of economic stability.

Pappus phenotypes and flight performance across evolutionary history in the daisy family.

Diversity in pappus shapes and size in Asteraceae suggests an adaptive response to dispersion challenges adjusting diaspore to optimal phenotypic configurations. Here, by analysing the relationship among pappus-cypsela size relationships, flight performance and pappus types in an evolutionary context, we evaluate the role of natural selection acting on the evolution of diaspore configuration at a macro-ecological scale in the daisy family. To link pappus-cypsela size relationships with flight performance we collected published data on these traits from 82 species. This allowed us to translate morphometric traits in flight performance for 150 species represented in a fully resolved backbone phylogeny of the daisy family. Through ancestral reconstructions and evolutionary model selection we assessed whether flight performance was associated with and constrained by different pappus types. Additionally, we evaluated, through phylogenetic regressions, whether species with different pappus types exhibited evolutionary allometric pappus-cypsela size relationships. The setose pappus type had the highest flight performances and represented the most probable ancestral state in the family. Stepwise changes in pappus types independently led from setose to multiple instances of pappus loss with associated reduction in flight performance. Flight performance evolution was best modelled as constrained by five adaptive regimes represented by specific pappus types which correspond with specific optimal diaspore configurations that are distinct in pappus-cypsela allometric relationships. Evolutionary modelling suggests natural selection as the main factor of diaspore configuration changes which proceeded towards five optima, often overcoming constraints imposed by allometric relationships and favouring evolution in certain directions. With the perspective that natural selection is the main process driving the observed patterns, various biotic and abiotic are suggested as principal drivers of transitions in diaspore configurations along space and time in the daisy family history. Results also allow discussion of evolutionary changes in a historical context.

Authenticity Verification of Alcoholic Beverages in IoT Environment through Fluorescence Hyperspectral Technology and Machine Learning

In the Internet of Things (IoT) environment, verifying the authenticity of beverages is an important issue. Baijiu, a unique Chinese distilled liquor, has gained popularity among people due to its distinctive flavor. In recent years, there has been increasing concern about the adulteration of baijiu. Many studies in this field have attempted to combine chemometrics, electronic nose, Raman spectroscopy, and other detection technologies with machine learning for baijiu detection. However, no research has yet applied the mature detection technology of fluorescence hyperspectral technology to this field. Additionally, existing studies have mainly focused on branded baijiu, without addressing the problem of adulteration in bulk baijiu, which remains a long-standing issue. In this study, a combination of fluorescence hyperspectral and machine learning was used for rapid and nondestructive adulteration detection of bulk liquor in Chengdu. The fluorescence hyperspectral data of three types of samples, namely, original liquor, adulterated liquor with water, and adulterated liquor with industrial alcohol, were obtained, and the raw data were preprocessed using Savitzky–Golay (SG), baseline correction (BL), multivariate scatter correction (MSC), and the combination of Savitzky–Golay and baseline correction (SG-BL). Finally, PCA and MDS are also used to extract spectral data characteristics, respectively. For model selection, three models were chosen: back-propagation artificial neural network (BP-ANN), support vector machine (SVM), and one-dimensional convolutional neural network (1D-CNN). For model evaluation, we choose accuracy, recall, precision, F1-score, memory occupied, and time consumed by the model to evaluate. The experiments show that the combination of fluorescence hyperspectral technology and machine learning can provide a new method, MSC-PCA-SVM, for bulk liquor detection, with a final classification accuracy, recall, precision, and F1-score all reaching 100%, The effectiveness and application potential of this method in the IoT environment have been demonstrated. The results of this study can not only be applied to liquor detection but also have the value of reference for the adulthood and detection of the entire food industry. It is of great significance to standardize the market and protect consumer rights.