期刊文献+
共找到47篇文章
< 1 2 3 >
每页显示 20 50 100
Selective ensemble modeling based on nonlinear frequency spectral feature extraction for predicting load parameter in ball mills 被引量:3
1
作者 汤健 柴天佑 +1 位作者 刘卓 余文 《Chinese Journal of Chemical Engineering》 SCIE EI CAS CSCD 2015年第12期2020-2028,共9页
Strong mechanical vibration and acoustical signals of grinding process contain useful information related to load parameters in ball mills. It is a challenge to extract latent features and construct soft sensor model ... Strong mechanical vibration and acoustical signals of grinding process contain useful information related to load parameters in ball mills. It is a challenge to extract latent features and construct soft sensor model with high dimensional frequency spectra of these signals. This paper aims to develop a selective ensemble modeling approach based on nonlinear latent frequency spectral feature extraction for accurate measurement of material to ball volume ratio. Latent features are first extracted from different vibrations and acoustic spectral segments by kernel partial least squares. Algorithms of bootstrap and least squares support vector machines are employed to produce candidate sub-models using these latent features as inputs. Ensemble sub-models are selected based on genetic algorithm optimization toolbox. Partial least squares regression is used to combine these sub-models to eliminate collinearity among their prediction outputs. Results indicate that the proposed modeling approach has better prediction performance than previous ones. 展开更多
关键词 Nonlinear latent feature extraction Kernel partial least squares selective ensemble modeling Least squares support vector machines Material to ball volume ratio
下载PDF
Ensemble habitat suitability modeling of stomatopods with Oratosquilla oratoria as an example
2
作者 Lisha Guan Xianshi Jin +1 位作者 Tao Yang Xiujuan Shan 《Acta Oceanologica Sinica》 SCIE CAS CSCD 2023年第4期93-102,共10页
Stomatopods are better known as mantis shrimp with considerable ecological importance in wide coastal waters globally. Some stomatopod species are exploited commercially, including Oratosquilla oratoria in the Northwe... Stomatopods are better known as mantis shrimp with considerable ecological importance in wide coastal waters globally. Some stomatopod species are exploited commercially, including Oratosquilla oratoria in the Northwest Pacific. Yet, few studies have published to promote accurate habitat identification of stomatopods, obstructing scientific management and conservation of these valuable organisms. This study provides an ensemble modeling framework for habitat suitability modeling of stomatopods, utilizing the O. oratoria stock in the Bohai Sea as an example. Two modeling techniques(i.e., generalized additive model(GAM) and geographical weighted regression(GWR)) were applied to select environmental predictors(especially the selection between two types of sediment metrics) that better characterize O. oratoria distribution and build separate habitat suitability models(HSM). The performance of the individual HSMs were compared on interpolation accuracy and transferability.Then, they were integrated to check whether the ensemble model outperforms either individual model, according to fishers’ knowledge and scientific survey data. As a result, grain-size metrics of sediment outperformed sediment content metrics in modeling O. oratoria habitat, possibly because grain-size metrics not only reflect the effect of substrates on burrow development, but also link to sediment heat capacity which influences individual thermoregulation. Moreover, the GWR-based HSM outperformed the GAM-based HSM in interpolation accuracy,while the latter one displayed better transferability. On balance, the ensemble HSM appeared to improve the predictive performance overall, as it could avoid dependence on a single model type and successfully identified fisher-recognized and survey-indicated suitable habitats in either sparsely sampled or well investigated areas. 展开更多
关键词 habitat suitability STOMATOPOD coastal fisheries predictor selection ensemble model
下载PDF
A New Selective Neural Network Ensemble Method Based on Error Vectorization and Its Application in High-density Polyethylene (HDPE) Cascade Reaction Process
3
作者 朱群雄 赵乃伟 徐圆 《Chinese Journal of Chemical Engineering》 SCIE EI CAS CSCD 2012年第6期1142-1147,共6页
Chemical processes are complex, for which traditional neural network models usually can not lead to satisfactory accuracy. Selective neural network ensemble is an effective way to enhance the generalization accuracy o... Chemical processes are complex, for which traditional neural network models usually can not lead to satisfactory accuracy. Selective neural network ensemble is an effective way to enhance the generalization accuracy of networks, but there are some problems, e.g., lacking of unified definition of diversity among component neural networks and difficult to improve the accuracy by selecting if the diversities of available networks are small. In this study, the output errors of networks are vectorized, the diversity of networks is defined based on the error vectors, and the size of ensemble is analyzed. Then an error vectorization based selective neural network ensemble (EVSNE) is proposed, in which the error vector of each network can offset that of the other networks by training the component networks orderly. Thus the component networks have large diversity. Experiments and comparisons over standard data sets and actual chemical process data set for production of high-density polyethylene demonstrate that EVSNE performs better in generalization ability. 展开更多
关键词 high-density polyethylene modeling selective neural network ensemble diversity definition error vectorization
下载PDF
Characterization of three-dimensional channel reservoirs using ensemble Kalman filter assisted by principal component analysis 被引量:2
4
作者 Byeongcheol Kang Hyungsik Jung +1 位作者 Hoonyoung Jeong Jonggeun Choe 《Petroleum Science》 SCIE CAS CSCD 2020年第1期182-195,共14页
Ensemble-based analyses are useful to compare equiprobable scenarios of the reservoir models.However,they require a large suite of reservoir models to cover high uncertainty in heterogeneous and complex reservoir mode... Ensemble-based analyses are useful to compare equiprobable scenarios of the reservoir models.However,they require a large suite of reservoir models to cover high uncertainty in heterogeneous and complex reservoir models.For stable convergence in ensemble Kalman filter(EnKF),increasing ensemble size can be one of the solutions,but it causes high computational cost in large-scale reservoir systems.In this paper,we propose a preprocessing of good initial model selection to reduce the ensemble size,and then,EnKF is utilized to predict production performances stochastically.In the model selection scheme,representative models are chosen by using principal component analysis(PCA)and clustering analysis.The dimension of initial models is reduced using PCA,and the reduced models are grouped by clustering.Then,we choose and simulate representative models from the cluster groups to compare errors of production predictions with historical observation data.One representative model with the minimum error is considered as the best model,and we use the ensemble members near the best model in the cluster plane for applying EnKF.We demonstrate the proposed scheme for two 3D models that EnKF provides reliable assimilation results with much reduced computation time. 展开更多
关键词 Channel reservoir CHARACTERIZATION MODEL selection scheme EGG MODEL Principal component analysis(PCA) ensemble KALMAN filter(EnKF) History matching
下载PDF
Optimized Ensemble Algorithm for Predicting Metamaterial Antenna Parameters 被引量:4
5
作者 El-Sayed M.El-kenawy Abdelhameed Ibrahim +3 位作者 Seyedali Mirjalili Yu-Dong Zhang Shaima Elnazer Rokaia M.Zaki 《Computers, Materials & Continua》 SCIE EI 2022年第6期4989-5003,共15页
Metamaterial Antenna is a subclass of antennas that makes use of metamaterial to improve performance.Metamaterial antennas can overcome the bandwidth constraint associated with tiny antennas.Machine learning is receiv... Metamaterial Antenna is a subclass of antennas that makes use of metamaterial to improve performance.Metamaterial antennas can overcome the bandwidth constraint associated with tiny antennas.Machine learning is receiving a lot of interest in optimizing solutions in a variety of areas.Machine learning methods are already a significant component of ongoing research and are anticipated to play a critical role in today’s technology.The accuracy of the forecast is mostly determined by the model used.The purpose of this article is to provide an optimal ensemble model for predicting the bandwidth and gain of the Metamaterial Antenna.Support Vector Machines(SVM),Random Forest,K-Neighbors Regressor,and Decision Tree Regressor were utilized as the basic models.The Adaptive Dynamic Polar Rose Guided Whale Optimization method,named AD-PRS-Guided WOA,was used to pick the optimal features from the datasets.The suggested model is compared to models based on five variables and to the average ensemble model.The findings indicate that the presented model using Random Forest results in a Root Mean Squared Error(RMSE)of(0.0102)for bandwidth and RMSE of(0.0891)for gain.This is superior to other models and can accurately predict antenna bandwidth and gain. 展开更多
关键词 Metamaterial antenna machine learning ensemble model feature selection guided whale optimization support vector machines
下载PDF
An Efficient Automated Technique for Classification of Breast Cancer Using Deep Ensemble Model
6
作者 Muhammad Zia Ur Rehman Jawad Ahmad +3 位作者 Emad Sami Jaha Abdullah Marish Ali Mohammed A.Alzain Faisal Saeed 《Computer Systems Science & Engineering》 SCIE EI 2023年第7期897-911,共15页
Breast cancer is one of the leading cancers among women.It has the second-highest mortality rate in women after lung cancer.Timely detection,especially in the early stages,can help increase survival rates.However,manu... Breast cancer is one of the leading cancers among women.It has the second-highest mortality rate in women after lung cancer.Timely detection,especially in the early stages,can help increase survival rates.However,manual diagnosis of breast cancer is a tedious and time-consuming process,and the accuracy of detection is reliant on the quality of the images and the radiologist’s experience.However,computer-aided medical diagnosis has recently shown promising results,leading to the need to develop an efficient system that can aid radiologists in diagnosing breast cancer in its early stages.The research presented in this paper is focused on the multi-class classification of breast cancer.The deep transfer learning approach has been utilized to train the deep learning models,and a pre-processing technique has been used to improve the quality of the ultrasound dataset.The proposed technique utilizes two deep learning models,Mobile-NetV2 and DenseNet201,for the composition of the deep ensemble model.Deep learning models are fine-tuned along with hyperparameter tuning to achieve better results.Subsequently,entropy-based feature selection is used.Breast cancer identification using the proposed classification approach was found to attain an accuracy of 97.04%,while the sensitivity and F1 score were 96.87%and 96.76%,respectively.The performance of the proposed model is very effective and outperforms other state-of-the-art techniques presented in the literature. 展开更多
关键词 Breast cancer image enhancement ensemble model transfer learning feature selection
下载PDF
Tissue specific prediction of N^(6)-methyladenine sites based on an ensemble of multi-input hybrid neural network
7
作者 CANGZHI JIA DONG JIN +1 位作者 XIN WANG QI ZHAO 《BIOCELL》 SCIE 2022年第4期1105-1121,共17页
N^(6)-Methyladenine is a dynamic and reversible post translational modification,which plays an essential role in various biological processes.Because of the current inability to identify m6A-containing mRNAs,computati... N^(6)-Methyladenine is a dynamic and reversible post translational modification,which plays an essential role in various biological processes.Because of the current inability to identify m6A-containing mRNAs,computational approaches have been developed to identify m6A sites in DNA sequences.Aiming to improve prediction performance,we introduced a novel ensemble computational approach based on three hybrid deep neural networks,including a convolutional neural network,a capsule network,and a bidirectional gated recurrent unit(BiGRU)with the self-attention mechanism,to identify m6A sites in four tissues of three species.Across a total of 11 datasets,we selected different feature subsets,after optimized from 4933 dimensional features,as input for the deep hybrid neural networks.In addition,to solve the deviation caused by the relatively small number of experimentally verified samples,we constructed an ensemble model through integrating five sub-classifiers based on different training datasets.When compared through 5-fold cross-validation and independent tests,our model showed its superiority to previous methods,im6A-TS-CNN and iRNA-m6A. 展开更多
关键词 M6A sites Deep hybrid neural networks ensemble model Feature selection
下载PDF
基于组合模型的复杂系统超多目标优化算法 被引量:1
8
作者 游雄雄 牛占文 《计算机集成制造系统》 EI CSCD 北大核心 2024年第4期1201-1212,共12页
代理模型辅助进化算法广泛用于昂贵的复杂工程系统优化设计,能够加速找到问题的最优解集。然而,单个模型预测性能依赖于具体问题,并且随着目标个数的增加,预测性能的不确定性将随之增加。因此,提出一种基于组合模型的复杂系统超多目标... 代理模型辅助进化算法广泛用于昂贵的复杂工程系统优化设计,能够加速找到问题的最优解集。然而,单个模型预测性能依赖于具体问题,并且随着目标个数的增加,预测性能的不确定性将随之增加。因此,提出一种基于组合模型的复杂系统超多目标优化算法。首先,建立组合代理模型并结合随机参考向量替代机制,以更好地搜索超多目标问题的非支配解集。其次,基于改进的统计下限最小值(LCB)准则及自适应个体选择策略选择优秀个体进行真实评估,以更新组合代理模型,使其能更好地辅助算法找到最优解集。最后,通过所提算法与已有代理模型进化算法在一系列测试函数和工程优化实例上的对比结果表明,所提算法具有良好的性能和潜力。 展开更多
关键词 超多目标优化 组合代理模型 统计下限最小值准则 个体选择策略
下载PDF
基于集成特征选择的中小微企业信贷风险分类模型研究 被引量:1
9
作者 路佳佳 王国兰 《中央民族大学学报(自然科学版)》 2024年第1期61-67,共7页
文章以客户违约率作为中小微企业信用风险的评价标准,尝试构造基于集成特征选择的中小微企业信用风险分类模型,结合互信息矩阵、基于k折交叉验证的随机森林和支持向量机对模型进行分析。研究表明企业的信誉等级、销项有效率和最高销项... 文章以客户违约率作为中小微企业信用风险的评价标准,尝试构造基于集成特征选择的中小微企业信用风险分类模型,结合互信息矩阵、基于k折交叉验证的随机森林和支持向量机对模型进行分析。研究表明企业的信誉等级、销项有效率和最高销项对信用风险有显著影响,其他因素对信用风险的影响不显著,实验说明基于k折交叉验证的支持向量机具有可靠的信贷风险预测能力,对中小微企业信用风险评估有较强的参考价值。 展开更多
关键词 集成特征选择 分类模型 支持向量机 信贷风险
下载PDF
基于进化集成学习的用户购买意向预测
10
作者 张一凡 于千城 张丽丝 《计算机应用研究》 CSCD 北大核心 2024年第2期368-374,共7页
在电子商务时代背景下,精准预测用户的购买意向已经成为提高销售效率和优化客户体验的关键因素。针对传统集成策略在模型设计阶段往往受人为因素限制的问题,构建了一种自适应进化集成学习模型用于预测用户的购买意向。该模型能够自适应... 在电子商务时代背景下,精准预测用户的购买意向已经成为提高销售效率和优化客户体验的关键因素。针对传统集成策略在模型设计阶段往往受人为因素限制的问题,构建了一种自适应进化集成学习模型用于预测用户的购买意向。该模型能够自适应地选择最优基学习器和元学习器,并融合基学习器的预测信息和特征间的差异性扩展特征维度,从而提高预测的准确性。此外,为进一步优化模型的预测效果,设计了一种二元自适应差分进化算法进行特征选择,旨在筛选出对预测结果有显著影响的特征。研究结果表明,与传统优化算法相比,二元自适应差分进化算法在全局搜索和特征选择方面表现优异。相较于六种常见的集成模型和DeepForest模型,所构建的进化集成模型在AUC值上分别提高了2.76%和2.72%,并且能够缓解数据不平衡所带来的影响。 展开更多
关键词 购买预测 差分进化算法 进化集成 特征选择 模型选择
下载PDF
基于SDL-LightGBM集成学习的软件缺陷预测模型
11
作者 谢华祥 高建华 黄子杰 《计算机工程与设计》 北大核心 2024年第3期769-776,共8页
为提高软件缺陷预测准确性和预测模型的可解释性,提出一种Spearman+DE+LIME+LightGBM(SDL-LightGBM)集成学习的软件缺陷预测模型。使用混合特征选择方法Spearman+LightGBM确定最佳特征子集,在保证模型预测性能的情况下降低模型复杂度;... 为提高软件缺陷预测准确性和预测模型的可解释性,提出一种Spearman+DE+LIME+LightGBM(SDL-LightGBM)集成学习的软件缺陷预测模型。使用混合特征选择方法Spearman+LightGBM确定最佳特征子集,在保证模型预测性能的情况下降低模型复杂度;使用集成学习算法LightGBM(light gradient boosting machine)对特征子集建立预测模型,并使用差分进化(differential evolution, DE)算法优化模型的重要超参数;使用局部可解释的模型无关技术(local interpretable model-agnostic explanations, LIME)对模型进行局部可解释分析。实验通过12个项目的35个版本的结果表明,SDL-LightGBM算法优于现有的软件缺陷预测方法,F1值平均提高8.97%,AUC值平均提高11.42%,模型训练时间缩短43.6%。 展开更多
关键词 缺陷预测 机器学习 集成学习 特征选择 模型优化 模型解释 差分进化
下载PDF
Selective Ensemble Extreme Learning Machine Modeling of Effluent Quality in Wastewater Treatment Plants 被引量:9
12
作者 Li-Jie Zhao 1,2 Tian-You Chai 2 De-Cheng Yuan 1 1 College of Information Engineering,Shenyang University of Chemical Technology,Shenyang 110042,China 2 State Key Laboratory of Synthetical Automation for Process Industries,Northeastern University,Shenyang 110189,China 《International Journal of Automation and computing》 EI 2012年第6期627-633,共7页
Real-time and reliable measurements of the effluent quality are essential to improve operating efficiency and reduce energy consumption for the wastewater treatment process.Due to the low accuracy and unstable perform... Real-time and reliable measurements of the effluent quality are essential to improve operating efficiency and reduce energy consumption for the wastewater treatment process.Due to the low accuracy and unstable performance of the traditional effluent quality measurements,we propose a selective ensemble extreme learning machine modeling method to enhance the effluent quality predictions.Extreme learning machine algorithm is inserted into a selective ensemble frame as the component model since it runs much faster and provides better generalization performance than other popular learning algorithms.Ensemble extreme learning machine models overcome variations in different trials of simulations for single model.Selective ensemble based on genetic algorithm is used to further exclude some bad components from all the available ensembles in order to reduce the computation complexity and improve the generalization performance.The proposed method is verified with the data from an industrial wastewater treatment plant,located in Shenyang,China.Experimental results show that the proposed method has relatively stronger generalization and higher accuracy than partial least square,neural network partial least square,single extreme learning machine and ensemble extreme learning machine model. 展开更多
关键词 Wastewater treatment process effluent quality prediction extreme learning machine selective ensemble model genetic algorithm.
原文传递
User Purchase Intention Prediction Based on Improved Deep Forest
13
作者 Yifan Zhang Qiancheng Yu Lisi Zhang 《Computer Modeling in Engineering & Sciences》 SCIE EI 2024年第4期661-677,共17页
Widely used deep neural networks currently face limitations in achieving optimal performance for purchase intention prediction due to constraints on data volume and hyperparameter selection.To address this issue,based... Widely used deep neural networks currently face limitations in achieving optimal performance for purchase intention prediction due to constraints on data volume and hyperparameter selection.To address this issue,based on the deep forest algorithm and further integrating evolutionary ensemble learning methods,this paper proposes a novel Deep Adaptive Evolutionary Ensemble(DAEE)model.This model introduces model diversity into the cascade layer,allowing it to adaptively adjust its structure to accommodate complex and evolving purchasing behavior patterns.Moreover,this paper optimizes the methods of obtaining feature vectors,enhancement vectors,and prediction results within the deep forest algorithm to enhance the model’s predictive accuracy.Results demonstrate that the improved deep forest model not only possesses higher robustness but also shows an increase of 5.02%in AUC value compared to the baseline model.Furthermore,its training runtime speed is 6 times faster than that of deep models,and compared to other improved models,its accuracy has been enhanced by 0.9%. 展开更多
关键词 Purchase prediction deep forest differential evolution algorithm evolutionary ensemble learning model selection
下载PDF
选择性融合多尺度筒体振动频谱的磨机负荷参数建模 被引量:14
14
作者 汤健 柴天佑 +2 位作者 丛秋梅 刘卓 余文 《控制理论与应用》 EI CAS CSCD 北大核心 2015年第12期1582-1591,共10页
针对目前采用经验模态分解(empirical model decomposition,EMD)得到的系列子信号构建的磨机负荷参数软测量模型泛化性能差、难以进行清晰物理解释,以及EMD算法存在的模态混叠等问题,本文提出了基于选择性融合多尺度筒体振动频谱的建模... 针对目前采用经验模态分解(empirical model decomposition,EMD)得到的系列子信号构建的磨机负荷参数软测量模型泛化性能差、难以进行清晰物理解释,以及EMD算法存在的模态混叠等问题,本文提出了基于选择性融合多尺度筒体振动频谱的建模方法.首先采用EMD、集合EMD(ensemble EMD,EEMD)、希尔伯特振动分解(Hilbert vibration decomposition,HVD)共3种多组分信号自适应分解算法获得磨机筒体振动多尺度子信号的集合,接着通过相关性分析剔除虚假无关部分,然后再将与原始信号相关性强的那部分多尺度子信号变换至频域,进而更有利于构建这些多尺度频谱与磨机负荷参数间的映射模型,最后通过改进分支定界选择性集成(improved branch and bound based selective ensemble,IBBSEN)算法建立软测量模型,实现对多源多尺度筒体振动频谱的最优选择性信息融合.基于实验球磨机运行数据的仿真实验表明所提方法在模型可解释性和泛化性能上均优于之前研究所提出方法. 展开更多
关键词 多组分信号分解 信息融合 选择性集成建模 振动频谱 软测量
下载PDF
选择性集成LTDGPR模型的自适应软测量建模方法 被引量:8
15
作者 熊伟丽 李妍君 《化工学报》 EI CAS CSCD 北大核心 2017年第3期984-991,共8页
随着时间的增加,传统时间差(TD)模型会出现性能显著下降的问题。为了提高TD模型的可靠性和预测精度,同时考虑过程的时滞特征,基于一种选择性集成策略,提出一种局部时间差高斯过程回归(LTDGPR)模型的自适应软测量建模方法。首先,提取出... 随着时间的增加,传统时间差(TD)模型会出现性能显著下降的问题。为了提高TD模型的可靠性和预测精度,同时考虑过程的时滞特征,基于一种选择性集成策略,提出一种局部时间差高斯过程回归(LTDGPR)模型的自适应软测量建模方法。首先,提取出数据库中的时滞动态信息,对建模数据进行重构;然后,采取局部化策略对差分后的重构样本进行统计划分,得到LTDGPR模型集。对于新来的输入样本,选择部分泛化能力强的LTDGPR模型进行集成,估计出含一定时间差的主导变量动态偏移值;最后,基于TD模型思想对当前时刻主导变量值进行在线预测。通过脱丁烷塔过程的数据建模仿真研究,验证了所提方法的有效性和精度。 展开更多
关键词 选择性集成 时间差模型 参数识别 动态建模 化学过程
下载PDF
基于选择性集成算法的浸出率混合预测模型 被引量:4
16
作者 胡广浩 毛志忠 杨菲 《仪器仪表学报》 EI CAS CSCD 北大核心 2013年第5期1049-1053,共5页
一个精确的模型对浸出过程中浸出率预测是十分重要的。针对湿法冶金浸出生产过程中浸出率在线检测的难点,提出一种有效的混合建模方法,建立浸出率的预测模型。在对浸出过程进行分析后,建立了一个浸出过程机理模型。由于机理模型与实际... 一个精确的模型对浸出过程中浸出率预测是十分重要的。针对湿法冶金浸出生产过程中浸出率在线检测的难点,提出一种有效的混合建模方法,建立浸出率的预测模型。在对浸出过程进行分析后,建立了一个浸出过程机理模型。由于机理模型与实际之间存在着较大的误差,因此建立了混合模型来减少误差。随后,针对小样本建模问题,提出了基于二进制PSO算法的选择性bagging集成算法,并将该算法应用于混合模型的误差补偿中去。实验结果表明该混合模型的预测精度比其他模型的预测精度高。 展开更多
关键词 浸出过程 预测 混合模型 选择性bagging集成算法
下载PDF
一种基于E-HMM的选择性集成人脸识别算法 被引量:2
17
作者 李金秀 高新波 +1 位作者 杨越 肖冰 《电子与信息学报》 EI CSCD 北大核心 2009年第2期288-292,共5页
基于嵌入式隐马尔可夫模型(Embedded Hidden Markov Model,E-HMM)的人脸识别方法的识别性能依赖于模型参数的合理选择。提出了一种基于E-HMM的多模型选择性集成人脸识别算法,选择出个体精度高且互补性强的模型来进行集成的人脸识别。实... 基于嵌入式隐马尔可夫模型(Embedded Hidden Markov Model,E-HMM)的人脸识别方法的识别性能依赖于模型参数的合理选择。提出了一种基于E-HMM的多模型选择性集成人脸识别算法,选择出个体精度高且互补性强的模型来进行集成的人脸识别。实验结果表明,与传统的基于E-HMM的人脸识别方法相比,新算法不仅可以获得更好、更稳定的识别效果,而且具有更强的泛化能力。 展开更多
关键词 人脸识别 嵌入式隐马尔可夫模型 模型选择 选择性集成 泛化能力
下载PDF
一种新的选择性神经网络集成方法及其在PTA中的应用 被引量:3
18
作者 朱群雄 孟庆浩 《化工学报》 EI CAS CSCD 北大核心 2009年第10期2510-2516,共7页
神经网络集成可以显著提高神经网络的泛化性能。传统的集成方法中大都采用将训练的所有网络直接进行组合的方式形成集成网络,而实际上这些网络可能具有一定的相关性。为此,选择性神经网络集成成为目前研究的热点,它能够进一步提高集成... 神经网络集成可以显著提高神经网络的泛化性能。传统的集成方法中大都采用将训练的所有网络直接进行组合的方式形成集成网络,而实际上这些网络可能具有一定的相关性。为此,选择性神经网络集成成为目前研究的热点,它能够进一步提高集成网络的泛化性能。本文提出了一种利用网络权值计算网络模型之间差异度的新的选择性神经网络集成方法DWSEN。UCI数据测试表明,与流行的集成方法Bagging和Boosting比较,本方法有着更好的泛化能力和稳定性。将DWSEN应用于精对苯二甲酸(PTA)溶剂系统脱水塔装置的建模过程,结果显示,利用该方法训练得到的集成模型具有更好的泛化性能,能够较好地模拟生产运行过程。 展开更多
关键词 神经网络集成 选择性集成 模型差异度 溶剂脱水塔
下载PDF
基于神经网络的建筑能耗混合预测模型 被引量:10
19
作者 于军琪 杨思远 +1 位作者 赵安军 高之坤 《浙江大学学报(工学版)》 EI CAS CSCD 北大核心 2022年第6期1220-1231,共12页
为了提升建筑能耗预测的精度、鲁棒性和泛化能力,提出树种算法(TSA)优化的径向基函数(RBF)神经网络与长短时记忆(LSTM)神经网络结合的混合预测模型.采用基于自适应噪声的完全集成经验模态分解算法,将建筑能耗数据分解为1组本征模态函数(... 为了提升建筑能耗预测的精度、鲁棒性和泛化能力,提出树种算法(TSA)优化的径向基函数(RBF)神经网络与长短时记忆(LSTM)神经网络结合的混合预测模型.采用基于自适应噪声的完全集成经验模态分解算法,将建筑能耗数据分解为1组本征模态函数(IMF)分量和1个残余分量,利用样本熵算法将各分量划分为高频分量和低频分量.采用最小绝对收缩与选择算子(LASSO)方法进行特征选择.分别利用TSA算法优化后的RBF模型与LSTM模型对低频分量和高频分量进行预测,并叠加重构得到最终预测结果.模型评估结果表明,混合预测模型的精度为98.72%.相比于RBF、TSA-RBF、LSTM模型,所提模型的预测效果更好,且具有较强的鲁棒性和泛化能力,能够更为有效地用于建筑逐时电力能耗预测. 展开更多
关键词 建筑能耗预测 神经网络 混合预测模型 集成经验模态分解 特征选择
下载PDF
基于局部能量的集成特征选择 被引量:2
20
作者 季薇 李云 《南京大学学报(自然科学版)》 CAS CSCD 北大核心 2012年第4期499-503,共5页
特征选择是机器学习和数据挖掘领域的关键问题之一,而特征选择的稳定性也是目前的一个研究热点.基于能量学习模型,分析了基于局部能量的特征选择方法并根据集成特征选择的原理,对基于局部能量的特征排序结果进行集成,以提高算法的稳定性... 特征选择是机器学习和数据挖掘领域的关键问题之一,而特征选择的稳定性也是目前的一个研究热点.基于能量学习模型,分析了基于局部能量的特征选择方法并根据集成特征选择的原理,对基于局部能量的特征排序结果进行集成,以提高算法的稳定性.在现实数据集上的实验结果表明集成特征选择可以有效提高算法的稳定性. 展开更多
关键词 特征选择 能量学习 集成
下载PDF
上一页 1 2 3 下一页 到第
使用帮助 返回顶部