In order to classify the minimal hepatic encephalopathy （MHE） patients from healthy controls, the independent component analysis （ICA） is used to generate the default mode network （DMN） from resting-state functi...In order to classify the minimal hepatic encephalopathy （MHE） patients from healthy controls, the independent component analysis （ICA） is used to generate the default mode network （DMN） from resting-state functional magnetic resonance imaging （fMRI）. Then a Bayesian voxel- wised method, graphical-model-based multivariate analysis （GAMMA）, is used to explore the associations between abnormal functional integration within DMN and clinical variable. Without any prior knowledge, five machine learning methods, namely, support vector machines （SVMs）, classification and regression trees （ CART ）, logistic regression, the Bayesian network, and C4.5, are applied to the classification. The functional integration patterns were alternative within DMN, which have the power to predict MHE with an accuracy of 98%. The GAMMA method generating functional integration patterns within DMN can become a simple, objective, and common imaging biomarker for detecting MIIE and can serve as a supplement to the existing diagnostic methods.展开更多
There are differences between the different individuals of learning. Adaptive learning support system is a learning system, which provides the learning supports suitable for the characteristics of the individuals acco...There are differences between the different individuals of learning. Adaptive learning support system is a learning system, which provides the learning supports suitable for the characteristics of the individuals according to the differences in the learning of individuals. In this paper, through the analysis on the adaptive learning support system, a system framework based on SOA is proposed and the research methods of the metadata model are emphatically discussed.展开更多
点云被广泛使用在各种三维应用场景中,但是实际应用中通常存在扫描、标注费时费力等局限性,因此基于小样本数据集的点云分类网络更加符合应用需求.为了有效地提高深度学习分类算法在小样本点云数据集上的分类效果,提出一种针对小样本数...点云被广泛使用在各种三维应用场景中,但是实际应用中通常存在扫描、标注费时费力等局限性,因此基于小样本数据集的点云分类网络更加符合应用需求.为了有效地提高深度学习分类算法在小样本点云数据集上的分类效果,提出一种针对小样本数据集的点云分类方法.针对训练数据集不平衡问题,首先采用基于相似度依赖的Dirichlet中餐馆过程对数据集进行预处理,在无需人工指定聚类个数的前提下对样本进行重新聚类,以提升分类网络在小样本数据集上的性能;然后在重新聚类后的样本上使用模型无关(model agnostic meta learning,MAML)算法训练PointNet++,达到用少量点云样本就能快速适应新任务的能力.所提方法不但降低了模型对数据量的依赖,提高了模型泛化能力,而且成功地把MAML算法从二维图像分类拓展到三维点云分类中;在Modelnet40数据集上的实验结果表明,与PointNet++相比,该方法的训练时间减少了一半,分类准确率平均提高6.67%,验证了该方法在小样本数据集上的有效性.展开更多
Building information modeling(BIM)object classification takes a lot of time and energy.Misclassification or omission of any object may lead to the emergence of abnormal results,which have a great impact on the project...Building information modeling(BIM)object classification takes a lot of time and energy.Misclassification or omission of any object may lead to the emergence of abnormal results,which have a great impact on the project workflow and results.Roundly understanding BIM object classification,by improving Swin Transformer classifier algorithm parameters,using the model primitives extracted from IFC format BIM model file,deep learning of 7 types of BIM object categories is taken.Through the performance and evaluation indicators obtained in training,the results improve the classification accuracy.展开更多
文摘针对共享电动汽车通过需求响应参与电力系统备用服务的可调度容量预测问题,基于历史轨迹数据提出一种基于模型无关的元学习(model-agnostic meta-learning,MAML)、卷积神经网络(convolutional neural network,CNN)、长短期记忆网络(long short term memory network,LSTM)和注意力机制(attention mechanism)的可调度容量评估模型,采用LSTM对CNN从历史数据中提取有效的特征向量动态变化进行建模学习,并用MAML对CNN-LSTM网络的初始化参数进行训练,在解决传统神经网络难以有效提取历史序列中潜在高维特征且当时序过长时重要信息易丢失的问题的同时,通过多任务训练对元预测网络进行微调以快速适应新预测任务,从而提高模型的预测精度及泛化能力;加入注意力机制突出对预测结果起关键性作用的时序信息,进一步提高预测精度。仿真结果表明所提模型可以有效预测不同日期类型和不同功能区域共享电动汽车的可调度容量,也为后续共享电动汽车通过需求响应参与电网备用服务的风险评估研究提供参考。
基金The National Natural Science Foundation of China(No.8123003481271739+2 种基金81501453)the Special Program of Medical Science of Jiangsu Province(No.BL2013029)the Natural Science Foundation of Jiangsu Province(No.BK20141342)
文摘In order to classify the minimal hepatic encephalopathy （MHE） patients from healthy controls, the independent component analysis （ICA） is used to generate the default mode network （DMN） from resting-state functional magnetic resonance imaging （fMRI）. Then a Bayesian voxel- wised method, graphical-model-based multivariate analysis （GAMMA）, is used to explore the associations between abnormal functional integration within DMN and clinical variable. Without any prior knowledge, five machine learning methods, namely, support vector machines （SVMs）, classification and regression trees （ CART ）, logistic regression, the Bayesian network, and C4.5, are applied to the classification. The functional integration patterns were alternative within DMN, which have the power to predict MHE with an accuracy of 98%. The GAMMA method generating functional integration patterns within DMN can become a simple, objective, and common imaging biomarker for detecting MIIE and can serve as a supplement to the existing diagnostic methods.
文摘There are differences between the different individuals of learning. Adaptive learning support system is a learning system, which provides the learning supports suitable for the characteristics of the individuals according to the differences in the learning of individuals. In this paper, through the analysis on the adaptive learning support system, a system framework based on SOA is proposed and the research methods of the metadata model are emphatically discussed.
文摘点云被广泛使用在各种三维应用场景中,但是实际应用中通常存在扫描、标注费时费力等局限性,因此基于小样本数据集的点云分类网络更加符合应用需求.为了有效地提高深度学习分类算法在小样本点云数据集上的分类效果,提出一种针对小样本数据集的点云分类方法.针对训练数据集不平衡问题,首先采用基于相似度依赖的Dirichlet中餐馆过程对数据集进行预处理,在无需人工指定聚类个数的前提下对样本进行重新聚类,以提升分类网络在小样本数据集上的性能;然后在重新聚类后的样本上使用模型无关(model agnostic meta learning,MAML)算法训练PointNet++,达到用少量点云样本就能快速适应新任务的能力.所提方法不但降低了模型对数据量的依赖,提高了模型泛化能力,而且成功地把MAML算法从二维图像分类拓展到三维点云分类中;在Modelnet40数据集上的实验结果表明,与PointNet++相比,该方法的训练时间减少了一半,分类准确率平均提高6.67%,验证了该方法在小样本数据集上的有效性.
文摘Building information modeling(BIM)object classification takes a lot of time and energy.Misclassification or omission of any object may lead to the emergence of abnormal results,which have a great impact on the project workflow and results.Roundly understanding BIM object classification,by improving Swin Transformer classifier algorithm parameters,using the model primitives extracted from IFC format BIM model file,deep learning of 7 types of BIM object categories is taken.Through the performance and evaluation indicators obtained in training,the results improve the classification accuracy.
