Effect of Brisk Walking on Self-Care Agency or Care Dependency among Colorectal Cancer Patients with Permanent Stoma

Objective:The purpose of this study was to determine the effectiveness of brisk walking as an intervention for self-care agency and care dependency in patients with permanent colorectal cancer stoma.Method:This study adopted a quasi-experimental research design,specifically a non-equivalent control group pre-test and post-test design.Utilizing the Exercise of Self-Care Agency Scale(ESCA)and Care Dependency Scale(CDS),a survey was administered to 64 patients from a hospital in Shandong Province.The statistical methods used for analyzing data included frequency,mean,standard deviation(SD),independent t-test,P-value calculation,and dependent t-test.Result:After two months of a brisk walking exercise program,participants in the experimental group had a higher level of self-care agency than before the experiment(P<0.05),and their level of care dependency was significantly reduced(P<0.05).Participants in the control group also showed higher levels of self-care agency(P<0.05)and lower levels of care dependency(P<0.05)after two months compared to their levels before the two months.Conclusion:The brisk walking program had a positive impact on patients’self-care agency and reduced their care dependency.

部分饱和岩石的黏弹性Chapman-Kelvin模型的正演模拟方法研究

在讨论地震波传播理论时,大部分情况下是把地震波看作弹性波.事实上,地下介质是非完全弹性介质,这使得传统的均匀、完全弹性介质理论受到了严重挑战.地震波在裂隙岩石中传播受裂隙系统和流体含量的影响较大,以往许多关于频变各向异性的理论大多局限于单相流体假设,但是几乎所有储层中通常被一种以上的流体部分饱和或完全饱和.本文在Chapman理论基础上,提出了基于部分饱和的黏弹性Chapman-Kelvin模型的正演模拟新方法,以提高对部分饱和岩石的黏弹性地震波频变各向异性的认识.该方法基于Chapman和Kelvin理论模型,计算了双相不混溶流体饱和裂隙岩石中的黏弹性波频变各向异性弹性系数的表达式,提出包含喷射流和斑块效应的统一地震波传播的黏弹性Chapman-Kelvin新模型.通过对新模型进行数值试验,讨论了无裂缝和存在裂缝两种情况下,含裂隙储层部分饱和岩石中耦合的喷射流和斑块效应对黏弹性介质频变各向异性的影响.试验结果反映出黏弹性介质下地震波频变各向异性的变化规律,验证了本文提出的新方法和新模型正确.本文将黏弹性介质各向异性与裂缝中的流体流动参数相联系,有利于提高对含裂缝储层部分饱和岩石的黏弹性波频变地震各向异性的认识,以及地震学与油藏工程的结合程度.

白藜芦醇通过AMPK/PPARα/PGC1α影响酒精依赖致肝损伤机制研究

目的 观察白藜芦醇(Resveratrol, RES)对酒精依赖大鼠肝组织损伤以及AMP依赖的蛋白激酶(adenosine 5-monophosphate-activated protein, AMPK)/过氧化物酶体增殖物激活受体α(peroxisome proliferator-activated receptorα,PPARα)/过氧化物酶体增殖物激活受体γ辅激活子1α(peroxisome proliferator activated receptor gamma coactivator 1 alpha, PGC1α)信号通路的影响,讨论RES治疗酒精依赖致肝损伤的修复机制,为临床应用RES提供科学依据。方法 将SPF级雄性大鼠采用随机数字表法设立空白对照组、酒精依赖模型组,采用双瓶自由饮建立20%的酒精依赖模型;酒精依赖模型组制备成功后随机分为酒精依赖模型组、RES高、中、低剂量治疗组,继续给与双瓶自由饮,RES按照100、50、25 mg/(kg·d)剂量灌胃14 d。建模期间检测大鼠体重、饮酒量和饮水量,RES治疗后蛋白免疫印迹法检测大鼠肝组织AMPK、磷酸化AMP依赖的蛋白激酶(phosphorylation-adenosine 5’-monophosphate-activated protein kinas, p-AMPK)、PPARα和PGC1α蛋白表达。结果 蛋白免疫印迹法结果显示:与空白对照组比较,酒精依赖模型组p-AMPK(t=6.61,P<0.05)、PPARα(t=3.08,P<0.05)和PGC1α(t=7.09,P<0.05)表达水平显著下调;与酒精依赖模型组比较,RES高剂量治疗组p-AMPK(F=8.60,P<0.05)、PPARα(F=4.38,P<0.05)和PGC1α(F=11.33,P<0.05)显著上调。结论 结果显示RES可能通过调控AMPK/PPARα/PGC1α信号发挥保护作用。

Long-term prognosis and its associated predictive factors in patients with eosinophilic gastroenteritis

BACKGROUND Eosinophilic gastroenteritis(EGE)is a chronic recurrent disease with abnormal eosinophilic infiltration in the gastrointestinal tract.Glucocorticoids remain the most common treatment method.However,disease relapse and glucocorticoid dependence remain notable problems.To date,few studies have illuminated the prognosis of EGE and risk factors for disease relapse.AIM To describe the clinical characteristics of EGE and possible predictive factors for disease relapse based on long-term follow-up.METHODS This was a retrospective cohort study of 55 patients diagnosed with EGE admitted to one medical center between 2013 and 2022.Clinical records were collected and analyzed.Kaplan-Meier curves and log-rank tests were conducted to reveal the risk factors for long-term relapse-free survival(RFS).RESULTS EGE showed a median onset age of 38 years and a slight female predominance(56.4%).The main clinical symptoms were abdominal pain(89.1%),diarrhea(61.8%),nausea(52.7%),distension(49.1%)and vomiting(47.3%).Forty-three(78.2%)patients received glucocorticoid treatment,and compared with patients without glucocorticoid treatments,they were more likely to have elevated serum immunoglobin E(IgE)(86.8%vs 50.0%,P=0.022)and descending duodenal involvement(62.8%vs 27.3%,P=0.046)at diagnosis.With a median follow-up of 67 mo,all patients survived,and 56.4%had at least one relapse.Six variables at baseline might have been associated with the overall RFS rate,including age at diagnosis<40 years[hazard ratio(HR)2.0408,95%confidence interval(CI):1.0082–4.1312,P=0.044],body mass index(BMI)>24 kg/m^(2)(HR 0.3922,95%CI:0.1916-0.8027,P=0.014),disease duration from symptom onset to diagnosis>3.5 mo(HR 2.4725,95%CI:1.220-5.0110,P=0.011),vomiting(HR 3.1259,95%CI:1.5246-6.4093,P=0.001),total serum IgE>300 KU/L at diagnosis(HR 0.2773,95%CI:0.1204-0.6384,P=0.022)and glucocorticoid treatment(HR 6.1434,95%CI:2.8446-13.2676,P=0.003).CONCLUSION In patients with EGE,younger onset age,longer disease course,vomiting and glucocorticoid treatment were risk factors for disease relapse,whereas higher BMI and total IgE level at baseline were protective.

融合多种时空自注意力机制的Transformer交通流预测模型

交通流预测是智能交通系统中实现城市交通优化的一种重要方法,准确的交通流量预测对交通管理和诱导具有重要意义.然而,因交通流本身存在高度时空依赖性而表现出复杂的非线性特征,现有的方法主要考虑路网中节点的局部时空特征,忽略了路网中所有节点的长期时空特征.为了充分挖掘交通流数据复杂的时空依赖,提出一种融合多种时空自注意力机制的Transformer交通流预测模型(MSTTF).该模型在嵌入层通过位置编码嵌入时间和空间信息,并在注意力机制层融合邻接空间自注意力机制,相似空间自注意力机制,时间自注意力机制,时间-空间自注意力机制等多种自注意力机制挖掘数据中潜在的时空依赖关系,最后在输出层进行预测.结果表明,MSTTF模型与传统时空Transformer相比,MAE平均降低了10.36%.特别地,相比于目前最先进的PDFormer模型,MAE平均降低了1.24%,能取得更好的预测效果.

Assessment of Dependent Performance Shaping Factors in SPAR-H Based on Pearson Correlation Coefficient

With the improvement of equipment reliability,human factors have become the most uncertain part in the system.The standardized Plant Analysis of Risk-Human Reliability Analysis(SPAR-H)method is a reliable method in the field of human reliability analysis(HRA)to evaluate human reliability and assess risk in large complex systems.However,the classical SPAR-H method does not consider the dependencies among performance shaping factors(PSFs),whichmay cause overestimation or underestimation of the risk of the actual situation.To address this issue,this paper proposes a new method to deal with the dependencies among PSFs in SPAR-H based on the Pearson correlation coefficient.First,the dependence between every two PSFs is measured by the Pearson correlation coefficient.Second,the weights of the PSFs are obtained by considering the total dependence degree.Finally,PSFs’multipliers are modified based on the weights of corresponding PSFs,and then used in the calculating of human error probability(HEP).A case study is used to illustrate the procedure and effectiveness of the proposed method.

生淀粉水解α-淀粉酶Amy486的重组表达和性质分析

为明晰生淀粉水解α-淀粉酶Amy486独特的催化特征,将克隆自Exiguobacterium sp.J84菌株的amy486进行重组表达,并研究重组酶的酶学性质、嗜盐特性和钙离子依赖性。Amy486最适催化pH为7.5,在pH 6.5~8.5范围内酶活力保持40%以上,最适温度为35℃,30℃半衰期为100 h。1.5 mol/L Na_(2)SO_(4)处理可将Amy486的比酶活由1.53 U/mg提升至2209 U/mg。添加1.0 mol/L Na_(2)SO_(4),Amy486在35℃放置500 h,酶活力可保持60%以上。2.5 mmol/L CaCl_(2)可提升酶活力至110%,添加超过5 mmol/L CaCl_(2),Amy486的相对酶活力降至100%以下,EDTA孵育对Amy486蛋白的酶活力和稳定性影响较小。Amy486与钙离子结合的关键位点为K_(3)0_(2),K_(3)0_(2)E与钙离子的结合能力增强,从而降低该酶对外源钙离子的依赖性。α-淀粉酶Amy486及其突变体酶K_(3)0_(2)E是具有较高比酶活的生淀粉水解酶,对外源钙离子的依赖性较低,其活力的发挥依赖于适合的盐浓度,可应用于某些高盐环境下的淀粉水解。

基于时空Transformer的多空间尺度交通预测模型

准确的交通预测对提高智能交通系统的效率至关重要。交通系统的空间依赖不仅体现在道路的相连关系上,更重要的是由道路属性、区域功能等因素形成的隐藏空间依赖。另外,交通数据之间的时间依赖具有严格的相对位置关系,忽略这一问题将难以实现准确的交通预测。为了解决这些问题,提出了一种基于时空Transformer的多空间尺度交通预测模型(MSS-STT)。MSS-STT使用多个特定的Transformer网络对不同的空间尺度建模,以捕捉隐藏空间依赖,同时使用图卷积网络来学习静态空间特征。接着,使用门控机制将不同空间尺度的空间依赖与静态空间特征根据各自对预测的重要性进行融合。最后,根据时间序列中不同相对位置对预测的不同贡献来提取不同的时间依赖关系。在PeMS数据集上的实验结果表明,MSS-STT优于最先进的基线。

Data Scale, Data Scope and Platform Enterprise Performance: Insights from Digital Platform M&As

Data is a key asset for digital platforms,and mergers and acquisitions(M&As)are an important way for platform enterprises to acquire it.The types of data obtained from intra-industry and cross-sector M&As differ,as does the extent to which they interact within or between platforms.The impact of such data on corporate market performance is an important question to consider when selecting strategies for digital platform M&As.Based on our research on advertising-driven platforms,we developed a two-stage Hotelling game model for comparing the market performance effects of intra-industry M&As and cross-sector M&As for digital platforms.We carried out an empirical test using relevant data from advertising-driven digital platforms between 2009 and 2021,as well as a case study on Baidu’s M&A activities.Our research discovered that intra-industry M&As driven by“data economies of scale”and cross-sector M&As driven by“data economies of scope”are both beneficial to the market performance of platform enterprises.Intra-industry M&As have a more significant positive effect on the market performance of platform enterprises because the same types of data are easier to integrate and develop the“network effect of data scale”.From a data factor perspective,this paper reveals the inherent economic logic by which different types of M&As influence the market performance of digital platforms,as well as policymaking recommendations for all digital platforms to select M&A strategies based on data scale,data scope,and the network effect of data.

Theoretical characterization of the temperature-dependent saturation magnetization of magnetic metallic materials

Based on the force-heat equivalence energy density principle,a theoretical model for magnetic metallic materials is developed,which characterizes the temperature-dependent magnetic anisotropy energy by considering the equivalent relationship between magnetic anisotropy energy and heat energy;then the relationship between the magnetic anisotropy constant and saturation magnetization is considered.Finally,we formulate a temperature-dependent model for saturation magnetization,revealing the inherent relationship between temperature and saturation magnetization.Our model predicts the saturation magnetization for nine different magnetic metallic materials at different temperatures,exhibiting satisfactory agreement with experimental data.Additionally,the experimental data used as reference points are at or near room temperature.Compared to other phenomenological theoretical models,this model is considerably more accessible than the data required at 0 K.The index included in our model is set to a constant value,which is equal to 10/3 for materials other than Fe,Co,and Ni.For transition metals(Fe,Co,and Ni in this paper),the index is 6 in the range of 0 K to 0.65T_(cr)(T_(cr) is the critical temperature),and 3 in the range of 0.65T_(cr) to T_(cr),unlike other models where the adjustable parameters vary according to each material.In addition,our model provides a new way to design and evaluate magnetic metallic materials with superior magnetic properties over a wide range of temperatures.

Review of Sc microalloying effects in Al-Cu alloys

Artificially controlling the solid-state precipitation in aluminum (Al) alloys is an efficient way to achieve well-performed properties,and the microalloying strategy is the most frequently adopted method for such a purpose.In this paper,recent advances in lengthscale-dependent scandium (Sc) microalloying effects in Al-Cu model alloys are reviewed.In coarse-grained Al-Cu alloys,the Sc-aided Cu/Sc/vacancies complexes that act as heterogeneous nuclei and Sc segregation at the θ′-Al_(2)Cu/matrix interface that reduces interfacial energy contribute significantly to θ′precipitation.By grain size refinement to the fine/ultrafine-grained scale,the strongly bonded Cu/Sc/vacancies complexes inhibit Cu and vacancy diffusing toward grain boundaries,promoting the desired intragranular θ′precipitation.At nanocrystalline scale,the applied high strain producing high-density vacancies results in the formation of a large quantity of (Cu Sc,vacancy)-rich atomic complexes with high thermal stability,outstandingly improving the strength/ductility synergy and preventing the intractable low-temperature precipitation.This review recommends the use of microalloying technology to modify the precipitation behaviors toward better combined mechanical properties and thermal stability in Al alloys.

Drug Use and Abuse among Orthopedic Ward Patients—A Revolving Cycle of Menace in Atbuth, Bauchi

Background: The use of drugs for purposes other than those for which they are meant to be used or in excess amounts. Psychoactive drugs are some of the drugs more commonly abused, also, antibiotics and other medications too can be misused. Drug abuse and misuse can lead to serious social, medical and emotional harm to the patients, and antibiotic resistance that makes treatment harder are also likely complications. Method: Patients in both male and female wards of the Orthoepaedics Department of ATBUTH, Bauchi were interviewed using a structured questionnaire and their responses were recorded and data were analyzed using the SPSS version 29. Results: A total of 112 patients were interviewed, 76 males and 36 females. Thirty-two (28.6%) patients had taken various kinds of unprescribed medications while on admission: 9 patients had taken unprescribed tramadol, 4 patients had taken codeine, 6 Ampiclox, 5 flagyl and about 8 patients had taken different kinds of traditional medications while on admission. Among the 32 patients, 23 (72%) are aged less than 30 years, 5 (16%) are aged between 30 and 50 years and 4 (12%) are aged above 50 years. Sixteen (50%) had such medications brought to them by relatives or friends, 10 (30%) were given by other patients on admission, 5 (15%) brought or bought the drugs by themselves while 1 (5%) were given by a non-clinical staff of the hospital. Conclusion: Drug misuse and abuse is a very serious, deleterious practice with destructive consequences in its wake, such consequences as drug dependency with all its antecedent effects, antibiotic resistance and difficulties in controlling/managing infections are but a few. So, it’s very important to both educate patients about these terrible practices and cope with the spread of them in our wards and hospitals.

Nonuniform Dependence on the Initial Data for Solutions of Conservation Laws

In this paper,we study systems of conservation laws in one space dimension.We prove that for classical solutions in Sobolev spaces H^(s),with s>3/2,the data-to-solution map is not uniformly continuous.Our results apply to all nonlinear scalar conservation laws and to nonlinear hyperbolic systems of two equations.

BCCLR:A Skeleton-Based Action Recognition with Graph Convolutional Network Combining Behavior Dependence and Context Clues

In recent years,skeleton-based action recognition has made great achievements in Computer Vision.A graph convolutional network(GCN)is effective for action recognition,modelling the human skeleton as a spatio-temporal graph.Most GCNs define the graph topology by physical relations of the human joints.However,this predefined graph ignores the spatial relationship between non-adjacent joint pairs in special actions and the behavior dependence between joint pairs,resulting in a low recognition rate for specific actions with implicit correlation between joint pairs.In addition,existing methods ignore the trend correlation between adjacent frames within an action and context clues,leading to erroneous action recognition with similar poses.Therefore,this study proposes a learnable GCN based on behavior dependence,which considers implicit joint correlation by constructing a dynamic learnable graph with extraction of specific behavior dependence of joint pairs.By using the weight relationship between the joint pairs,an adaptive model is constructed.It also designs a self-attention module to obtain their inter-frame topological relationship for exploring the context of actions.Combining the shared topology and the multi-head self-attention map,the module obtains the context-based clue topology to update the dynamic graph convolution,achieving accurate recognition of different actions with similar poses.Detailed experiments on public datasets demonstrate that the proposed method achieves better results and realizes higher quality representation of actions under various evaluation protocols compared to state-of-the-art methods.

An Improved CREAM Model Based on DS Evidence Theory and DEMATEL

Cognitive Reliability and Error Analysis Method(CREAM)is widely used in human reliability analysis(HRA).It defines nine common performance conditions(CPCs),which represent the factors thatmay affect human reliability and are used to modify the cognitive failure probability(CFP).However,the levels of CPCs are usually determined by domain experts,whichmay be subjective and uncertain.What’smore,the classicCREAMassumes that the CPCs are independent,which is unrealistic.Ignoring the dependence among CPCs will result in repeated calculations of the influence of the CPCs on CFP and lead to unreasonable reliability evaluation.To address the issue of uncertain information modeling and processing,this paper introduces evidence theory to evaluate the CPC levels in specific scenarios.To address the issue of dependence modeling,the Decision-Making Trial and Evaluation Laboratory(DEMATEL)method is used to process the dependence among CPCs and calculate the relative weights of each CPC,thus modifying the multiplier of the CPCs.The detailed process of the proposed method is illustrated in this paper and the CFP estimated by the proposed method is more reasonable.

Temperature inversion enables superior stability for low-temperature Zn-ion batteries

It is challenging for aqueous Zn-ion batteries(ZIBs)to achieve comparable low-temperature(low-T)performance due to the easy-frozen electrolyte and severe Zn dendrites.Herein,an aqueous electrolyte with a low freezing point and high ionic conductivity is proposed.Combined with molecular dynamics simulation and multi-scale interface analysis(time of flight secondary ion mass spectrometry threedimensional mapping and in-situ electrochemical impedance spectroscopy method),the temperature independence of the V_(2)O_(5)cathode and Zn anode is observed to be opposite.Surprisingly,dominated by the solvent structure of the designed electrolyte at low temperatures,vanadium dissolution/shuttle is significantly inhibited,and the zinc dendrites caused by this electrochemical crosstalk are greatly relieved,thus showing an abnormal temperature inversion effect.Through the disclosure and improvement of the above phenomena,the designed Zn||V_(2)O_(5)full cell delivers superior low-T performance,maintaining almost 99%capacity retention after 9500 cycles(working more than 2500 h)at-20°C.This work proposes a kind of electrolyte suitable for low-T ZIBs and reveals the inverse temperature dependence of the Zn anode,which might offer a novel perspective for the investigation of low-T aqueous battery systems.

Graph Convolutional Networks Embedding Textual Structure Information for Relation Extraction

Deep neural network-based relational extraction research has made significant progress in recent years,andit provides data support for many natural language processing downstream tasks such as building knowledgegraph,sentiment analysis and question-answering systems.However,previous studies ignored much unusedstructural information in sentences that could enhance the performance of the relation extraction task.Moreover,most existing dependency-based models utilize self-attention to distinguish the importance of context,whichhardly deals withmultiple-structure information.To efficiently leverage multiple structure information,this paperproposes a dynamic structure attention mechanism model based on textual structure information,which deeplyintegrates word embedding,named entity recognition labels,part of speech,dependency tree and dependency typeinto a graph convolutional network.Specifically,our model extracts text features of different structures from theinput sentence.Textual Structure information Graph Convolutional Networks employs the dynamic structureattention mechanism to learn multi-structure attention,effectively distinguishing important contextual features invarious structural information.In addition,multi-structure weights are carefully designed as amergingmechanismin the different structure attention to dynamically adjust the final attention.This paper combines these featuresand trains a graph convolutional network for relation extraction.We experiment on supervised relation extractiondatasets including SemEval 2010 Task 8,TACRED,TACREV,and Re-TACED,the result significantly outperformsthe previous.

Rural Resilience: A Comprehensive Study on Water Supply, Sanitation, Disease Patterns, and Hygiene Practices in Munshiganj, Bangladesh

This research project investigates the current status of water supply, sanitation, and hygiene practices in Munshiganj District, Bangladesh. Data collection involved a structured questionnaire and a reconnaissance survey. Findings reveal that 30% of individuals rely on surface water (hand-tube wells, rivers, and ponds), prioritized as canal > river > pond, while 70% depend on groundwater (subterranean electric motor, deep tube-well). Drinking water is generally sufficient, with 95% reporting adequacy throughout the year. About 45% use hand tube-well water, 28% use deep tube-well water, and 11% use supply tap water for various purposes. Bathing trends include underground water through electric motor > pond > hand tube-well water > river, while for cooking, the order is underground water through electric motor > pond > hand tube-well water > river. Toilet water supply ranks as supply tap water > hand tube-well water > deep tube-well water. Although sanitation awareness is high, some lack knowledge of good hygiene practices. After defecating, handwashing methods include soap, ash, soil, or water. Children’s waste disposal varies, with some discarding it in open areas. Approximately 40% suffer from diseases like Diarrhoea due to unsafe water, primarily affecting children and elders. Training exists, but a significant portion lacks sanitation education. Dry skin or exposure to cold water may cause temporary irritation. Local government involvement in sanitation efforts is less active compared to non-governmental organizations. Results emphasize the need to enhance community awareness of safe water supplies and sanitation practices. .

Extension of Flow Behaviour and Damage Models for Cast Iron Alloys with Strain Rate Effect

Cast iron alloys with low production cost and quite good mechanical properties are widely used in the automotive industry.To study the mechanical behavior of a typical ductile cast iron(GJS-450)with nodular graphite,uni-axial quasi-static and dynamic tensile tests at strain rates of 10^(-4),1,10,100,and 250 s^(-1)were carried out.In order to investigate the influence of stress state on the deformation and fracture parameters,specimens with various geometries were used in the experiments.Stress strain curves and fracture strains of the GJS-450 alloy in the strain rate range of 10^(-4)to 250 s^(-1)were obtained.A strain rate-dependent plastic flow model was proposed to describe the mechanical behavior in the corresponding strain-rate range.The available damage model was extended to take the strain rate into account and calibrated based on the analysis of local fracture strains.Simulations with the proposed plastic flow model and the damage model were conducted to observe the deformation and fracture process.The results show that the strain rate has obviously nonlinear effects on the yield stress and fracture strain of GJS-450 alloys.The predictions with the proposed plastic flow and damage models at various strain rates agree well with the experimental results,which illustrates that the rate-dependent plastic flow and damage models can be used to describe the mechanical behavior of cast iron alloys at elevated strain rates.The proposed plastic flow and damage models can be used to describe the deformation and fracture analysis of materials with similar properties.

Predicting Rock Burst in Underground Engineering Leveraging a Novel Metaheuristic-Based LightGBM Model

Rock bursts represent a formidable challenge in underground engineering,posing substantial risks to both infrastructure and human safety.These sudden and violent failures of rock masses are characterized by the rapid release of accumulated stress within the rock,leading to severe seismic events and structural damage.Therefore,the development of reliable prediction models for rock bursts is paramount to mitigating these hazards.This study aims to propose a tree-based model—a Light Gradient Boosting Machine(LightGBM)—to predict the intensity of rock bursts in underground engineering.322 actual rock burst cases are collected to constitute an exhaustive rock burst dataset,which serves to train the LightGBMmodel.Two population-basedmetaheuristic algorithms are used to optimize the hyperparameters of the LightGBM model.Finally,the sensitivity analysis is used to identify the predominant factors that may incur the occurrence of rock bursts.The results show that the population-based metaheuristic algorithms have a good ability to search out the optimal hyperparameters of the LightGBM model.The developed LightGBM model yields promising performance in predicting the intensity of rock bursts,with which accuracy on training and testing sets are 0.972 and 0.944,respectively.The sensitivity analysis discloses that the risk of occurring rock burst is significantly sensitive to three factors:uniaxial compressive strength(σc),stress concentration factor(SCF),and elastic strain energy index(Wet).Moreover,this study clarifies the particular impact of these three factors on the intensity of rock bursts through the partial dependence plot.