检索结果-内蒙古大学图书馆

Microblog Dimensionality Reduction-A deep Learning Approach

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2016年第7期28卷 1779-1789页

作者： Xu, Lei Jiang, Chunxiao Ren, Yong Chen, Hsiao-Hwa Tsinghua Univ Dept Comp Sci & Technol Beijing 100084 Peoples R China Tsinghua Univ Dept Elect Engn Beijing 100084 Peoples R China Natl Cheng Kung Univ Dept Engn Sci Tainan 70101 Taiwan

Exploring potentially useful information from huge amount of textual data produced by microblogging services has attracted much attention in recent years. An important preprocessing step of microblog text mining is to convert natural language texts into proper numerical representations. Due to the short-length characteristics of microblog texts, using term frequency vectors to represent microblog texts will cause "sparse data" problem. Finding proper representations of microblog texts is a challenging issue. In this paper, we apply deep networks to map the high-dimensional representations of microblog texts to low-dimensional representations. To improve the result of dimensionality reduction, we take advantage of the semantic similarity derived from two types of microblog-specific information, namely the retweet relationship and hashtags. Two types of approaches, including modifying training data and modifying the training objective of deep networks, are proposed to make use of microblog-specific information. Experiment results show that the deep models perform better than traditional dimensionality reduction methods such as latent semantic analysis and latent Dirichlet allocation topic model, and the use of microblog-specific information can help to learn better representations.

关键词： Microblog mining dimension reduction text representation semantic relatedness deep autoencoder

来源：评论

学校读者我要写书评

暂无评论

Dynamic Predictive Maintenance Scheduling Using deep Learning Ensemble for System Health Prognostics

引用

IEEE SENSORS JOURNAL 2021年第23期21卷 26878-26891页

作者： Chen, Chuang Zhu, Zheng Hong Shi, Jiantao Lu, Ningyun Jiang, Bin Nanjing Univ Aeronaut & Astronaut Coll Automat Engn Nanjing 211106 Peoples R China York Univ Lassonde Sch Engn Toronto ON M3J 1P3 Canada Nanjing Res Inst Elect Technol Nanjing 210039 Peoples R China

Modern engineering systems are usually equipped with a variety of sensors to measure real-time operating conditions. Based on the condition monitoring data from multiple sensor sources, this paper deals with a dynamic predictive maintenance scheduling using deep learning ensemble for system health prognostics. The deep learning ensemble model is composed of deep autoencoder and bidirectional long short-term memory in series, and aims to accurately estimate the system health state and remaining useful life. The deep autoencoder is used to extract the deep representative features hidden in condition monitoring data, whereas the inclusion of the bidirectional long short-term memory allows learning the temporal correlation information of features in both forward and backward time directions. Thus, the combination of the two models forms an effective model. With obtained prognostic information, optimal maintenance decisions are determined by two designed rules. The first rule presets a reliable remaining useful life threshold to dynamically judge whether maintenance is carried out at current inspection time, while the second one selects a degradation state to dynamically determine whether to place a spare part order. The performance of the proposed dynamic predictive maintenance strategy is measured by the maintenance cost per unit operating time (cost rate) using the aero-engine dataset from NASA. Its effectiveness is demonstrated by comparing with recent publications.

关键词： Predictive maintenance Sensors Feature extraction Predictive models Degradation Data mining Condition monitoring Predictive maintenance health state assessment remaining useful life prediction deep autoencoder bidirectional long short-term memory maintenance cost rate

来源：评论

学校读者我要写书评

暂无评论

Separation of Metabolites and Macromolecules for Short-TE ¹H-MRSI Using Learned Component-Specific Representations

引用

IEEE TRANSACTIONS ON MEDICAL IMAGING 2021年第4期40卷 1157-1167页

作者： Li, Yahang Wang, Zepeng Sun, Ruoyu Lam, Fan Univ Illinois Dept Bioengn Urbana IL 61801 USA Univ Illinois Dept Ind & Enterprise Syst Engn Urbana IL 61801 USA Univ Illinois Beckman Inst Adv Sci & Technol Urbana IL 61801 USA

Short-echo-time (TE) proton magnetic resonance spectroscopic imaging (MRSI) allows for simultaneously mapping a number of molecules in the brain, and has been recognized as an important tool for studying in vivo biochemistry in various neuroscience and disease applications. However, separation of the metabolite and macromolecule (MM) signals present in the short-TE data with significant spectral overlaps remains a major technical challenge. This work introduces a new approach to solve this problem by integrating imaging physics and representation learning. Specifically, a mixed unsupervised and supervised learning-based strategy was developed to learn the metabolite and MM-specific low-dimensional representations using deep autoencoders. A constrained reconstruction formulation is proposed to integrate the MRSI spatiospectral encoding model and the learned representations as effective constraints for signal separation. An efficient algorithm was developed to solve the resulting optimization problem with provable convergence. Simulation and experimental results have been obtained to demonstrate the component-specific representation power of the learned models and the capability of the proposed method in separating metabolite and MM signals for practical short-TE H-1-MRSI data.

关键词： Proton (H-1) magnetic resonance spectroscopic imaging short TE signal separation deep learning deep autoencoder low-dimensional models

来源：评论

学校读者我要写书评

暂无评论

A robust unsupervised anomaly detection framework

引用

APPLIED INTELLIGENCE 2022年第6期52卷 6022-6036页

作者： Luo, Zhengyu He, Kejing Yu, Zhixing South China Univ Technol Guangzhou Guangdong Peoples R China

Anomaly detection plays an essential role in monitoring dependable systems and networks such as computer clusters, water treatment systems, sensor networks, etc. However, anomaly detection nowadays remains a big challenge since previous researches suffer from inaccessible anomaly labels and inconsistent data types. Therefore, we propose a robust unsupervised anomaly detection framework (RUAD) to tackle the above problems. RUAD combines a deep autoencoder and a robust layer to extract the latent representations of data and separate normal data from abnormal data respectively, then utilizes Gaussian Mixture Model (GMM) to learn the distribution of normal data. In addition, our model can adapt to different types of data by simply modifying the structure of the deep autoencoder. Extensive experiments show that RUAD outperforms state-of-art anomaly detection techniques.

关键词： Robust unsupervised anomaly detection deep autoencoder Robust layer GMM

来源：评论

学校读者我要写书评

暂无评论

Hybrid Attention and Motion Constraint for Anomaly Detection in Crowded Scenes

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2023年第5期33卷 2259-2274页

作者： Zhang, Xinfeng Fang, Jinpeng Yang, Baoqing Chen, Shuhan Li, Bin Yangzhou Univ Coll Artificial Intelligence Coll Informat Engn Jiangsu Prov Engn Res Ctr Knowledge Management Yangzhou 225127 Jiangsu Peoples R China

Crowds often appear in surveillance videos in public places, from which anomaly detection is of great importance to public safety. Since the abnormal cases are rare, variable and unpredictable, autoencoders with encoder and decoder structures using only normal samples have become a hot topic among various approaches for anomaly detection. However, since autoencoders have excessive generalization ability, they can sometimes still reconstruct abnormal cases very well. Recently, some researchers construct memory modules under normal conditions and use these normal memory items to reconstruct test samples during inference to increase the reconstruction errors for anomalies. However, in practice, the errors of reconstructing normal samples with the memory items often increase as well, which makes it still difficult to distinguish between normal and abnormal cases. In addition, the memory-based autoencoder is usually available only in the specific scene where the memory module is constructed and almost loses the prospect of cross-scene applications. We mitigate the overgeneralization of autoencoders from a different perspective, namely, by reducing the prediction errors for normal cases rather than increasing the prediction errors for abnormal cases. To this end, we propose an autoencoder based on hybrid attention and motion constraint for anomaly detection. The hybrid attention includes the channel attention used in the encoding process and spatial attention added to the skip connection between the encoder and decoder. The hybrid attention is introduced to reduce the weight of the feature channels and regions representing the background in the feature matrix, which makes the autoencoder features more focused on optimizing the representation of the normal targets during training. Furthermore, we introduce motion constraint to improve the autoencoder's ability to predict normal activities in crowded scenes. We conduct experiments on real-world surveillance videos, UCSD, CU

关键词： Videos Anomaly detection Training Memory modules Dictionaries Testing Surveillance video surveillance deep autoencoder attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Developing deep learning based regression approaches for determination of chemical compositions in dry black goji berries (Lycium ruthenicum Murr.) using near-infrared hyperspectral imaging

引用

FOOD CHEMISTRY 2020年 319卷 126536-000页

作者： Zhang, Chu Wu, Wenyan Zhou, Lei Cheng, Huan Ye, Xingqian He, Yong Zhejiang Univ Coll Biosyst Engn & Food Sci Hangzhou 310058 Peoples R China Minist Agr & Rural Affairs Key Lab Spect Sensing Hangzhou 310058 Peoples R China Zhejiang Univ Natl Local Joint Engn Lab Intelligent Food Techno Zhejiang Key Lab Agrofood Proc Zhejiang Engn Lab Food Technol & Equipment Hangzhou 310058 Peoples R China Zhejiang Univ Fuli Inst Food Sci Hangzhou 310058 Peoples R China Zhejiang Univ Ningbo Res Inst Ningbo 315100 Peoples R China

Black goji berry (Lycium ruthenicum Murr.) has great commercial and nutritional values. Near-infrared hyper-spectral imaging (NIR-HSI) was used to determine total phenolics, total flavonoids and total anthocyanins in dry black goji berries. Convolutional neural networks (CNN) were designed and developed to predict the chemical compositions. These CNN models and deep autoencoder were used as supervised and unsupervised feature extraction methods, respectively. Partial least squares (PLS) and least-squares support vector machine (LS-SVM) as modelling methods, successive projections algorithm and competitive adaptive reweighted sampling (CARS) as wavelength selection methods, and principal component analysis (PCA) and wavelet transform (WT) as feature extraction methods were studied as conventional approaches for comparison. deep learning approaches as modelling methods and feature extraction methods obtained good and equivalent performances to the conventional methods. The results illustrated that deep learning had great potential as modelling and feature extraction methods for chemical compositions determination in NIR-HSI.

关键词： Black goji berry Near-infrared hyperspectral imaging Convolutional neural network deep autoencoder Regression issue Total phenolics Total flavonoids Total anthocyanins

来源：评论

学校读者我要写书评

暂无评论

Squeezing bottlenecks: Exploring the limits of autoencoder semantic representation capabilities

引用

NEUROCOMPUTING 2016年第PartB期175卷 1001-1008页

作者： Gupta, Parth Banchs, Rafael E. Rosso, Paolo Univ Politecn Valencia PRHLT Res Ctr E-46022 Valencia Spain Inst Infocomm Res Singapore Singapore

We present a comprehensive study on the use of autoencoders for modelling text data, in which (differently from previous studies) we focus our attention on the various issues. We explore the suitability of two different models binary deep autencoders (bDA) and replicated-softmax deep autencoders (rsDA) for constructing deep autoencoders for text data at the sentence level. We propose and evaluate two novel metrics for better assessing the text-reconstruction capabilities of autoencoders. We propose an automatic method to find the critical bottleneck dimensionality for text representations (below which structural information is lost);and finally we conduct a comparative evaluation across different languages, exploring the regions of critical bottleneck dimensionality and its relationship to language perplexity. (C) 2015 Elsevier B.V. All rights reserved.

关键词： Text representation deep autoencoder

来源：评论

学校读者我要写书评

暂无评论

deep discriminative dictionary pair learning for image classification

引用

APPLIED INTELLIGENCE 2023年第19期53卷 22017-22030页

作者： Zhu, Wenjie Peng, Bo Chen, Chunchun Chen, Hao China Jiliang Univ Coll Informat Engn Key Lab Electromagnet Wave Informat Technol & Metr Hangzhou 310018 Zhejiang Peoples R China Univ Queensland St Lucia Brisbane Qld 4072 Australia China Jiliang Univ Coll Sci Hangzhou 310018 Zhejiang Peoples R China

Discriminative dictionary learning has been extensively used for pattern classification tasks. By incorporating different kinds of label information into the dictionary learning framework, a dictionary can be attained that represents the original signal with discriminative reconstruction. The previous works learn the dictionary in the original space which limits the dictionary learning performance. In this paper, we propose an approach, namely deep Discriminative Dictionary Pair Learning ((DPL)-P-3) for image classification. The input of (DPL)-P-3 is not the matrix collected by original gray images or hand-crafted features but the relatively deeper features derived from autoencoders. Then, a structured dictionary is designed based on the discriminative contributions across different classes to reconstruct the deep feature. In addition, the associated structured projective dictionary is learned as well to guarantee the decoders updating towards the minimal error of deconvolution operator. By leveraging the discriminative-dictionary-learning-based loss function and the autoencoder loss function, (DPL)-P-3 can simultaneously learn the deep potential feature and the corresponding dictionary pair. In the testing phase of (DPL)-P-3, the minimum error between the deep feature and the structured projective component with regard to different classes can directly indicate the label by a basic matrix multiplication operation. Experimental results on challenging Extended Yale B, AR, UMIST, COIL20, Scene 15, and Caltech101 datasets demonstrate that the proposed (DPL)-P-3 outperforms the prominent dictionary learning methods.

关键词： Dictionary learning deep autoencoder Projective dictionary Dictionary pair learning Image classification

来源：评论

学校读者我要写书评

暂无评论

Predictive anomaly detection for marine diesel engine based on echo state network and autoencoder

引用

ENERGY REPORTS 2022年 8卷 998-1003页

作者： Qu, Chong Zhou, Zhiguo Liu, Zhiwen Jia, Shuli Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Shanghai Marine Diesel Engine Res Inst Automat Engn Dept Shanghai 201108 Peoples R China

Marine diesel engine with high thermal efficiency and good economy has become the main power of ships. Anomaly detection is an important method to improve the operation reliability of marine diesel engine. Most anomaly detection research focuses on failures that have occurred, and few studies consider anomaly prediction. A predictive anomaly detection method based on echo state network (ESN) and deep autoencoder is proposed. Historical sample data is collected and used to train the prediction network ESN and the anomaly detection network deep auto-encoder. After training, the prediction network ESN is used to predict the sensor data sequence in the future. And the predicted sequence is input into the anomaly detection network deep auto-encoder to obtain the predictive anomaly detection result. The relative error and root mean square error of the proposed method are at least 0.089 and 1.002 lower than other methods, respectively. Compared with other anomaly detection methods, the proposed autoencoder method obtains the best precision, accurate, recall indicators. Experiments show that it is feasible to establish a predictive anomaly detection method. More experiments under different conditions need to be studied, and higher performance algorithms need to be developed in the future. (C) 2022 Published by Elsevier Ltd.

关键词： Anomaly detection Prediction Marine diesel engine Echo state network deep autoencoder

来源：评论

学校读者我要写书评

暂无评论

Research on unsupervised anomaly data detection method based on improved automatic encoder and Gaussian mixture model

引用

JOURNAL OF CLOUD COMPUTING-ADVANCES SYSTEMS AND APPLICATIONS 2022年第1期11卷 1-16页

作者： Liu, Xiangyu Zhu, Shibing Yang, Fan Liang, Shengjun Univ Space Engn Beijing 101416 Peoples R China Beijing Informat Sci & Technol Univ Beijing 100026 Peoples R China

With the development of cloud computing, more and more security problems like "fuzzy boundary" are exposed. To solve such problems, unsupervised anomaly detection is increasingly used in cloud security, where density estimation is commonly used in anomaly detection clustering tasks. However, in practical use, the excessive amount of data and high dimensionality of data features can lead to difficulties in data calibration, data redundancy, and reduced effectiveness of density estimation algorithms. Although auto-encoders have made fruitful progress in data dimensionality reduction, using auto-encoders alone may still cause the model to be too generalized and unable to detect specific anomalies. In this paper, a new unsupervised anomaly detection method, MemAe-gmm-ma, is proposed. MemAe-gmm-ma generates a low-dimensional representation and reconstruction error for each input sample by a deep auto-encoder. It adds a memory module inside the auto-encoder to better learn the inner meaning of the training samples, and finally puts the low-dimensional information of the samples into a Gaussian mixture model (GMM) for density estimation. MemAe-gmm-ma demonstrates better performance on the public benchmark dataset, with a 4.47% improvement over the MemAe model standard F1 score on the NSL-KDD dataset, and a 9.77% improvement over the CAE-GMM model standard F1 score on the CIC-IDS-2017 dataset.

关键词： Cloud security Unsupervised machine learning Anomalous data detection Memory module deep autoencoder Gaussian mixture model

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：