Clinical outcome and prognostic factors of T4N0M0 colon cancer after R0 resection:A retrospective study

BACKGROUND Paradoxically,patients with T4N0M0(stage II,no lymph node metastasis)colon cancer have a worse prognosis than those with T2N1-2M0(stage III).However,no previous report has addressed this issue.AIM To screen prognostic risk factors for T4N0M0 colon cancer and construct a prognostic nomogram model for these patients.METHODS Two hundred patients with T4N0M0 colon cancer were treated at Tianjin Medical University General Hospital between January 2017 and December 2021,of which 112 patients were assigned to the training cohort,and the remaining 88 patients were assigned to the validation cohort.Differences between the training and validation groups were analyzed.The training cohort was subjected to multi-variate analysis to select prognostic risk factors for T4N0M0 colon cancer,followed by the construction of a nomogram model.RESULTS The 3-year overall survival(OS)rates were 86.2%and 74.4%for the training and validation cohorts,respectively.Enterostomy(P=0.000),T stage(P=0.001),right hemicolon(P=0.025),irregular review(P=0.040),and carbohydrate antigen 199(CA199)(P=0.011)were independent risk factors of OS in patients with T4N0M0 colon cancer.A nomogram model with good concordance and accuracy was constructed.CONCLUSION Enterostomy,T stage,right hemicolon,irregular review,and CA199 were independent risk factors for OS in patients with T4N0M0 colon cancer.The nomogram model exhibited good agreement and accuracy.

In-situ measurement via the flow-through method and numerical simulations for radon exhalation during measurements of the radon exhalation rate

Small-scale measurements of the radon exhalation rate using the flow-through and closed-loop methods were conducted on the surface of a uranium tailing pond to better understand the differences between the two methods.An abnormal radon exhalation behavior was observed,leading to computational fluid dynamics(CFD)-based simulations in which dynamic radon migration in a porous medium and accumulation chamber was considered.Based on the in-situ experimental and numerical simulation results,variations in the radon exhalation rate subject to permeability,flow rate,and insertion depth were quantified and analyzed.The in-situ radon exhalation rates measured using the flow-through method were higher than those measured using the closed-loop method,which could be explained by the negative pressure difference between the inside and outside of the chamber during the measurements.The consistency of the variations in the radon exhalation rate between the experiments and simulations suggests the reliability of CFD-based techniques in obtaining the dynamic evolution of transient radon exhalation rates for diffusion and convection at the porous medium-air interface.The synergistic effects of the three factors(insertion depth,flow rate,and permeability)on the negative pressure difference and measured exhalation rate were quantified,and multivariate regression models were established,with positive correlations in most cases;the exhalation rate decreased with increasing insertion depth at a permeability of 1×10^(−11) m^(2).CFD-based simulations can provide theoretical guidance for improving the flow-through method and thus achieve accurate measurements.

Multivariate Analysis of Female Stress Urinary Incontinence and Establishment of a Prediction Model

Stress urinary incontinence(SUI)is a symptom of uncontrolled urine outflow that affects millions of women worldwide[1].SUI is a significant healthcare issue that affects the quality of life of women across numerous domains,including social activities,physical health,mental well-being,employment,and sexual life.

Weathering and accumulation of trace elements in the soils of the Porali Plain, Balochistan: repercussions in agriculture

This study is thefirst attempt to assess the nature of the soil,especially on the western side of the Porali Plain in Balochistan;a new emerging agriculture hub,using weathering and pollution indices supplemented by multi-variate analysis based on geochemical data.The outcomes of this study are expected to help farmers in soil manage-ment and selecting suitable crops for the region.Twenty-five soil samples were collected,mainly from the arable land of the Porali Plain.After drying and coning-quarter-ing,soil samples were analyzed for major and trace ele-ments using the XRF technique;sieving and hydrometric methods were employed for granulometric analysis.Esti-mated data were analyzed using Excel,SPSS,and Surfer software to calculate various indices,correlation matrix,and spatial distribution.The granulometric analysis showed that 76%of the samples belonged to loam types of soil,12%to sand type,and 8%to silt type.Weathering indices:CIA,CIW,PIA,PWI,WIP,CIX,and ICV were calculated to infer the level of alteration.These indices reflect mod-erate to intense weathering;supported by K_(2)O/AI_(2)O_(3),Rb/K_(2)O,Rb/Ti,and Rb/Sr ratios.Assessment of the geo-ac-cumulation and Nemerow Pollution indices pinpoint rela-tively high concentrations of Pb,Ni,and Cr concentration in the soils.The correlation matrix and Principal Compo-nent Analysis show that the soil in this study area is mainly derived from the weathering of igneous rocks of Bela Ophiolite(Cretaceous age)and Jurassic sedimentary rocks of Mor Range having SEDEX/MVT type mineralization.Weathering may result in the undesirable accumulation of certain trace elements which adversely affects crops.

Geochemical Anomalies Identified by Multifractal Modeling: Implications for Mineral Exploration in the Ziyoutun Cu-Au District, Jilin Province, China

The Ziyoutun Cu-Au district is located in the Jizhong–Yanbian Metallogenic Belt and possesses excellent prospects. However, the thick regolith and complex tectonic settings present challenges in terms of detecting and decomposition of weak geochemical anomalies. To address this challenge, we initially conducted a comprehensive analysis of 1:10,000-scale soil geochemical data. This analysis included multivariate statistical techniques, such as correlation analysis, R-mode cluster analysis, Q–Q plots and factor analysis. Subsequently, we decomposed the geochemical anomalies, identifying weak anomalies using spectrum-area modeling and local singularity analysis. The results indicate that the assemblage of Au-Cu-Bi-As-Sb represents the mineralization at Ziyoutun. In comparison to conventional methods, spectrumarea modeling and local singularity analysis outperform in terms of identification of anomalies. Ultimately, we considered four specific target areas(AP01, AP02, AP03 and AP04) for future exploration, based on geochemical anomalies and favorable geological factors. Within AP01 and AP02, the geochemical anomalies suggest potential mineralization at depth, whereas in AP03 and AP04 the surface anomalies require additional geological investigation. Consequently, we recommend conducting drilling, following more extensive surface fieldwork, at the first two targets and verifying surface anomalies in the last two targets. We anticipate these findings will significantly enhance future exploration in Ziyoutun.

A Novel Graph Structure Learning Based Semi-Supervised Framework for Anomaly Identification in Fluctuating IoT Environment

With the rapid development of Internet of Things(IoT)technology,IoT systems have been widely applied in health-care,transportation,home,and other fields.However,with the continuous expansion of the scale and increasing complexity of IoT systems,the stability and security issues of IoT systems have become increasingly prominent.Thus,it is crucial to detect anomalies in the collected IoT time series from various sensors.Recently,deep learning models have been leveraged for IoT anomaly detection.However,owing to the challenges associated with data labeling,most IoT anomaly detection methods resort to unsupervised learning techniques.Nevertheless,the absence of accurate abnormal information in unsupervised learning methods limits their performance.To address these problems,we propose AS-GCN-MTM,an adaptive structural Graph Convolutional Networks(GCN)-based framework using a mean-teacher mechanism(AS-GCN-MTM)for anomaly identification.It performs better than unsupervised methods using only a small amount of labeled data.Mean Teachers is an effective semi-supervised learning method that utilizes unlabeled data for training to improve the generalization ability and performance of the model.However,the dependencies between data are often unknown in time series data.To solve this problem,we designed a graph structure adaptive learning layer based on neural networks,which can automatically learn the graph structure from time series data.It not only better captures the relationships between nodes but also enhances the model’s performance by augmenting key data.Experiments have demonstrated that our method improves the baseline model with the highest F1 value by 10.4%,36.1%,and 5.6%,respectively,on three real datasets with a 10%data labeling rate.

Multivariate Time Series Anomaly Detection Based on Spatial-Temporal Network and Transformer in Industrial Internet of Things

In the Industrial Internet of Things(IIoT),sensors generate time series data to reflect the working state.When the systems are attacked,timely identification of outliers in time series is critical to ensure security.Although many anomaly detection methods have been proposed,the temporal correlation of the time series over the same sensor and the state(spatial)correlation between different sensors are rarely considered simultaneously in these methods.Owing to the superior capability of Transformer in learning time series features.This paper proposes a time series anomaly detection method based on a spatial-temporal network and an improved Transformer.Additionally,the methods based on graph neural networks typically include a graph structure learning module and an anomaly detection module,which are interdependent.However,in the initial phase of training,since neither of the modules has reached an optimal state,their performance may influence each other.This scenario makes the end-to-end training approach hard to effectively direct the learning trajectory of each module.This interdependence between the modules,coupled with the initial instability,may cause the model to find it hard to find the optimal solution during the training process,resulting in unsatisfactory results.We introduce an adaptive graph structure learning method to obtain the optimal model parameters and graph structure.Experiments on two publicly available datasets demonstrate that the proposed method attains higher anomaly detection results than other methods.

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other fields.Furthermore,it is important to construct a digital twin system.However,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted accuracy.In this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)layer.Specifically,we fuse the spatial-temporal graph based on the interrelationship of spatial graphs.Simultaneously,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding methods.Subsequently,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)module.The module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like data.AFSTGCN dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted accuracy.Experiments on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

Multivariate analysis of oral mucosal ulcers during orthodontic treatment

BACKGROUND Orthodontic treatment can easily cause local soft tissue reactions in the oral cavity of patients under mechanical stress,leading to oral mucosal ulcers and affecting their quality of life.At present,only limited literature has explored the factors leading to oral ulcers in orthodontic treatment,and these research results are still controversial.AIM To investigate the current status and related factors of oral mucosal ulcers during orthodontic treatment,aiming to provide a valuable reference for preventing this disease in clinical practice.METHODS A total of 587 patients who underwent orthodontic treatment at the Peking University School of Stomatology and Hospital of Stomatology between 2020 and 2022 were selected and allocated to an observation or control group according to the incidence of oral mucosal ulcers during orthodontic therapy.A questionnaire survey was constructed to collect patient data,including basic information,lifestyle and eating habits,treatment details,mental factors,and trace element levels,and a comparative analysis of this data was performed between the two groups.RESULTS A logistic regression model with oral ulcers as the dependent variable was established.The regression results showed that age(≥60 years:odds ratio[OR]:6.820;95%confidence interval[CI]:2.226–20.893),smoking history(smoking:OR:4.434;95%CI:2.527–7.782),toothbrush hardness(hard:OR:2.804;95%CI:1.746–4.505),dietary temperature(hot diet:OR:1.399;95%CI:1.220–1.722),treatment course(>1 year:OR:3.830;95%CI:2.203–6.659),and tooth brushing frequency(>1 time per day:OR:0.228;95%CI:0.138–0.377)were independent factors for oral mucosal ulcers(P<0.05).Furthermore,Zn level(OR:0.945;95%CI:0.927–0.964)was a protective factor against oral ulcers,while the SAS(OR:1.284;95%CI:1.197–1.378)and SDS(OR:1.322;95%CI:1.231–1.419)scores were risk factors.CONCLUSION Age≥60 years,smoking history,hard toothbrush,hot diet,treatment course for>1 year,tooth brushing frequency of≤1 time per day,and mental anxiety are independent risk factors for oral mucosal ulcers.Therefore,these factors should receive clinical attention and be incorporated into the development and optimization of preventive strategies for reducing oral ulcer incidence.

Analysis of optical axis deviation caused by structural stiffness in equatorial telescopes

The impact of structural stiffness on optical axis deviation poses a significant challenge in the design of equatorial telescope structures.A comprehensive analysis during the design process can reduce the reliance of a telescope on advanced control technologies,thereby improving its economic feasibility.Although full-system finite element analyses are reliable,they are encumbered by significant time requirements and limitations in covering all possible telescope orientations.Therefore,we propose an efficient and comprehensive analytical method to evaluate the optical axis deviation of equatorial telescopes across a full range of angles.To address the challenge of ensuring that the analysis covers all possible positions of an equatorial telescope,based on a model from SiTian project,we analyze the optical axis deviations caused by the fork arm at 25 different angles and then use fitting methods to obtain results for all angles.Based on the analysis results of the optical axis deviation caused by the stiffness of the optical tube in the horizontal position,we derive the results for the tube at any position using geometric relationships.Finally,we calculate the coupling factors and combine these impacts.Furthermore,we identify six discrete feature points to reflect possible telescope orientations and conduct comprehensive finite element analyses.The results are in alignment with those acquired through a comprehensive computational approach.

Socioeconomic traits and the risk of Barrett’s esophagus and gastroesophageal reflux disease: A Mendelian randomization study

BACKGROUND Previous observational studies have shown that the prevalence of gastroesophageal reflux disease(GERD)and Barrett’s esophagus(BE)is associated with socioeconomic status.However,due to the methodological limitations of traditional observational studies,it is challenging to definitively establish causality.AIM To explore the causal relationship between the prevalence of these conditions and socioeconomic status using Mendelian randomization(MR).METHODS We initially screened single nucleotide polymorphisms(SNPs)to serve as proxies for eight socioeconomic status phenotypes for univariate MR analysis.The inverse variance weighted(IVW)method was used as the primary analytical method to estimate the causal relationship between the eight socioeconomic status phenotypes and the risk of GERD and BE.We then collected combinations of SNPs as composite proxies for the eight socioeconomic phenotypes to perform multivariate MR(MVMR)analyses based on the IVW MVMR model.Furthermore,a two-step MR mediation analysis was used to examine the potential mediation of the associations by body mass index,major depressive disorder(MDD),smoking,alcohol consumption,and sleep duration.RESULTS The study identified three socioeconomic statuses that had a significant impact on GERD.These included household income[odds ratio(OR):0.46;95% confidence interval(95%CI):0.31-0.70],education attainment(OR:0.23;95%CI:0.18-0.29),and the Townsend Deprivation Index at recruitment(OR:1.57;95%CI:1.04-2.37).These factors were found to independently and predominantly influence the genetic causal effect of GERD.Furthermore,the mediating effect of educational attainment on GERD was found to be mediated by MDD(proportion mediated:10.83%).Similarly,the effect of educational attainment on BE was mediated by MDD(proportion mediated:10.58%)and the number of cigarettes smoked per day(proportion mediated:3.50%).Additionally,the mediating effect of household income on GERD was observed to be mediated by sleep duration(proportion mediated:9.75%)CONCLUSION This MR study shed light on the link between socioeconomic status and GERD or BE,providing insights for the prevention of esophageal cancer and precancerous lesions.

Research on Total Electric Field Prediction Method of Ultra-High Voltage Direct Current Transmission Line Based on Stacking Algorithm

Ultra-high voltage(UHV)transmission lines are an important part of China’s power grid and are often surrounded by a complex electromagnetic environment.The ground total electric field is considered a main electromagnetic environment indicator of UHV transmission lines and is currently employed for reliable long-term operation of the power grid.Yet,the accurate prediction of the ground total electric field remains a technical challenge.In this work,we collected the total electric field data from the Ningdong-Zhejiang±800 kV UHVDC transmission project,as of the Ling Shao line,and perform an outlier analysis of the total electric field data.We show that the Local Outlier Factor(LOF)elimination algorithm has a small average difference and overcomes the performance of Density-Based Spatial Clustering of Applications with Noise(DBSCAN)and Isolated Forest elimination algorithms.Moreover,the Stacking algorithm has been found to have superior prediction accuracy than a variety of similar prediction algorithms,including the traditional finite element.The low prediction error of the Stacking algorithm highlights the superior ability to accurately forecast the ground total electric field of UHVDC transmission lines.

Quality Evaluation of Cardamine macrophylla Willd as an Edible and Medicinal Herb of Tibetan and Qiang Ethnic Groups

[Objectives]To evaluate the quality of Cardamine macrophylla Willd as Tibetan and Qiang medicinal materials,so as to improve its quality standard and evaluate the quality of C.macrophylla Willd in western Sichuan Province.[Methods]C.macrophylla Willd produced from western Sichuan Province was used as the sample,and the contents of moisture,total ash,acid-insoluble ash,extract,total flavonoids and quercetin in the ground part of C.macrophylla Willd were determined in accordance with the methods of Chinese Pharmacopoeia(2020 edition).With the above seven indicators as evaluation indicators,the quality of medicinal materials was comprehensively evaluated by cluster analysis and principal component analysis(PCA).[Results]According to the results of each indicator,the moisture content of C.macrophylla Willd sample should not exceed 11.00%,the total ash content should not exceed 18%,the acid-insoluble ash content should not exceed 6%,the extract content should not be less than 19%,the total flavone content(calculated by quercetin)should not be less than 2%,and the quercetin content should not be less than 0.15%.[Conclusions]The sample S7 has the best quality and S6 has the worst quality.In this study,the quantitative analysis method of total flavonoids(quercetin)and quercetin in C.macrophylla Willd was established,and the limits of each indicator were preliminarily formulated.

Accuracy Enhancement of the Folin-Ciocalteu Method for Propolis

This study presents an optimization of the Folin-Ciocalteu spectrophotometric method for the determination of total phenol content. Multivariate optimization using factorial planning 22 with a central point and central composite planning was constructed to evaluate the influence of variables in the process and maximize radiation absorption with minimal radiation scattering caused by solid formation. X-ray fluorescence and X-ray diffraction spectrometry were used to evaluate the chemical composition of solids formed and nephelometric and spectrophotometric studies were also used to evaluate whether the type, origin, dilution and dry extract contents of commercial propolis extracts would significantly influence the increase in radiation scattering and absorption. The optimized methodology added several advantages, such as reduction of reagents, time analysis, and higher accuracy.

Temporal and spatial variations hydrochemical components and driving factors in Baiyangdian Lake in the Northern Plain of China

Understanding the temporal and spatial variation of hydrochemical components in large freshwater lakes is crucial for effective management and conversation.In this study,we identify the temporalspatial characteristics and driving factors of the hydrochemical components in Baiyangdian Lake using geochemical methods(Gibbs diagram,Piper diagram and End-element diagram of ion ratio)and multivariate statistical techniques(Principal component analysis and Correlation analysis).16 sets of samples were collected from Baiyangdian Lake in May(normal season),July(flood season),and December(dry season)of 2022.Results indicate significant spatial variation in Nat,ci,SO and NO,,suggesting a strong influence of human activities.Cation concentrations exhibit greater seasonal variation in the dry season compared to the flood season,while the concentrations of the four anions show inconsistent seasonal changes due to the combined effects of river water chemical composition and human activities.The hydrochemical type of Baiyangdian Lake is primarily HCO,Cl-Na.Ca,Mg*and HCO,originate mainly from silicate and carbonate rock dissolution,while Kt,Nat and CI originate mainly from sewage and salt dissolution in sediments.SO42 may mainly stem from industrial wastewater,while NO,primarily originates from animal feces and domestic sewage.Through the use of Principal Component Analysis,it is identified that water-rock interaction(silicate and carbonate rocks dissolution,and dissolution of salt in sediments),carbonate sedimentation,sewage,agricultural fertilizer and manure,and nitrification are the main driving factors of the variation of hydrochemical components of Baiyangdian Lake across three hydrological seasons.These findings suggest the need for effective control of substandard domestic sewage discharge,optimization of agricultural fertilization strategies,and proper management of animal manure to comprehensively improve the water environment in Baiyangdian Lake.

Characteristics and genesis of groundwater salinization in coastal areas of the Lower Reaches of Oujiang Basin

The coastal areas of the lower reaches of Oujiang River Basin are rich in groundwater resources.However,the unsustainable exploitation and utilization of groundwater have led to significant changes in the groundwater environment.Understanding the characteristics and genesis of groundwater salinization is crucial for preventing its deterioration and ensuring sustainable utilization.In this study,a comprehensive approach combining the ion ratio method,mineral saturation index method and multivariate statistical analysis was employed to investigate the hydrochemical characteristics and main controlling factors in the study area.The findings reveal that:(1)Groundwater samples in study area exhibit a neutral to slightly alkaline pH.The predominant chemical types of unconfined water are HCO_(3)-Ca·Na,HCO_(3)·Cl-Na·Ca and HCO_(3)·SO_(4)-Ca·Na,while confined water mainly exhibits Cl·HCO_(3)-Na and Cl-Na types.(2)Salinity coefficients indicate an increase in salinity from unconfined to confined water.TDS,Na^(+)and Cl^(–)concentrations show an increasing trend from mountainous to coastal areas in unconfined water,while confined water displays variability in TDS,Na^(+)and Cl^(–)concentrations.(3)Groundwater salinity is mainly influenced by water-rock interactions,including the dissolution of halite and gypsum,cation exchange,and seawater intrusion etc.Additionally,human activities and carbonate dissolution contribute to salinity in unconfined water.Seawater intrusion is identified as the primary factor leading to higher salinity in confined water compared to unconfined water,with increasing cation exchange and seawater interaction observed from unconfined to confined water.

R-Factor Analysis of Data Based on Population Models Comprising R- and Q-Factors Leads to Biased Loading Estimates

Effects of performing an R-factor analysis of observed variables based on population models comprising R- and Q-factors were investigated. Although R-factor analysis of data based on a population model comprising R- and Q-factors is possible, this may lead to model error. Accordingly, loading estimates resulting from R-factor analysis of sample data drawn from a population based on a combination of R- and Q-factors will be biased. It was shown in a simulation study that a large amount of Q-factor variance induces an increase in the variation of R-factor loading estimates beyond the chance level. Tests of the multivariate kurtosis of observed variables are proposed as an indicator of possible Q-factor variance in observed variables as a prerequisite for R-factor analysis.

Distribution, Mobility, and Health Risks Assessment of Trace Metals in River Sediments from Intense Agricultural Activity Areas in West Africa

The economy of West African countries is mainly based on agriculture. However, the trace metal(loid)s contamination status in rivers is relatively unknown in the region. In this work, 45 surface sediments collected from the Bandama, Comoé, and Bia Rivers in south and south eastern Côte d’Ivoire (West Africa), were analyzed for total metal concentrations and chemical speciation. The results showed that the river sediments were considerably contaminated by Cd and moderately contaminated by As, Cu, Pb, and Zn. Significant spatial variations were observed among the stations but not between the rivers. Metals Cd and Cu were likely to cause more ecological risks. The speciation analysis unravelled that the metal(loid)s partitioned mainly in the residual fraction, with the potential mobile fraction varying from 14% to 28%. The study calls for establishment of strict policies relative to the application of fertilizers and agrochemicals and mining activities to protect the environment and human health risks.

Volatile profile and multivariant analysis of Sanhuang chicken breast in combination with Chinese 5-spice blend and garam masala

Sanhuang chicken is a popular native breed in China and well-known for delicious flavour.Spices could enhance the chicken meat flavour and work well in preservation.Chinese 5-spice blend(CS)and garam masala(GM)are routinely using spices in China and Pakistan,respectively.The flavour profiles of Sanhuang chicken breast(CB)and its blends with CS and GM were obtained by electronic nose(E-nose),solid-phase microextraction gas chromatography-mass spectrometry(SPME GC-MS)and GC-ion mobility spectrometry(GC-IMS).Principal component analysis(PCA)efficiently discriminated the aroma profiles of three chicken formulations.The GC-chromatographs revealed the significant aroma alterations of chicken breast meat after marination with spices.Aldehydes were the major contributors of chicken aroma,while most of the aromatic hydrocarbons were generated by spices.Almost all chicken key-compounds produced by oxidation reaction were either reduced or eliminated by marination,showing the antioxidation capacity of spices leading to meat preservation.GC-IMS is not only a rapid and comprehensive detection method,but also proved to be more sensitive than GC-MS.The substantial role of both traditional spices in enhancing flavour quality of chicken meat,and their exposure as functional ingredients in Chinese and Pakistan cuisines could lead to the cross-cultural meat trade opportunities.

Data-driven intelligent modeling framework for the steam cracking process

Steam cracking is the dominant technology for producing light olefins,which are believed to be the foundation of the chemical industry.Predictive models of the cracking process can boost production efficiency and profit margin.Rapid advancements in machine learning research have recently enabled data-driven solutions to usher in a new era of process modeling.Meanwhile,its practical application to steam cracking is still hindered by the trade-off between prediction accuracy and computational speed.This research presents a framework for data-driven intelligent modeling of the steam cracking process.Industrial data preparation and feature engineering techniques provide computational-ready datasets for the framework,and feedstock similarities are exploited using k-means clustering.We propose LArge-Residuals-Deletion Multivariate Adaptive Regression Spline(LARD-MARS),a modeling approach that explicitly generates output formulas and eliminates potentially outlying instances.The framework is validated further by the presentation of clustering results,the explanation of variable importance,and the testing and comparison of model performance.