检索结果-内蒙古大学图书馆

Social movie recommender system based on deep autoencoder network using Twitter data

NEURAL COMPUTING & APPLICATIONS 2021年第5期33卷 1607-1623页

作者： Tahmasebi, Hossein Ravanmehr, Reza Mohamadrezaei, Rezvan Islamic Azad Univ Dept Comp Engn Cent Tehran Branch Tehran Iran

Recommender systems attempt to provide effective suggestions to each user based on their interests and behaviors. These recommendations usually match the personal user preferences and assist them in the decision-making process. With the ever-expanding growth of information on the web, online education systems, e-commerce, and, eventually, the emergence of social networks, the necessity of developing such systems is unavoidable. Collaborative filtering and content-based filtering are among the most important techniques used in recommender systems. Meanwhile, with the significant advances in deep learning in recent years, the use of this technology has been widely observed in recommender systems. In this study, a hybrid social recommender system utilizing a deep autoencoder network is introduced. The proposed approach employs collaborative and content-based filtering, as well as users' social influence. The social influence of each user is calculated based on his/her social characteristics and behaviors on Twitter. For the evaluation purpose, the required datasets have been collected from MovieTweetings and Open Movie Database. The evaluation results show that the accuracy and effectiveness of the proposed approach have been improved compared to the other state-of-the-art methods.

关键词： Social recommender system deep learning deep autoencoder network Collaborative filtering Content-based filtering Social influence

来源：评论

学校读者我要写书评

暂无评论

Infomax-based deep autoencoder network for recognition of multi-element geochemical anomalies linked to mineralization

引用

COMPUTERS & GEOSCIENCES 2023年第1期175卷

作者： Esmaeiloghli, Saeid Tabatabaei, Seyed Hassan Carranza, Emmanuel John M. Isfahan Univ Technol Dept Min Engn Esfahan *** Iran Univ Free State Dept Geol ZA-9301 Bloemfontein South Africa

In recent years, deep autoencoder networks (DANs) have shown enormous potential to achieve state-of-the-art performance for recognizing multi-element geochemical anomalies related to mineralization. By training a DAN, multi-element signatures of geochemical background are learned by higher-level representations of input signals, providing key references to quantify reconstruction errors linked to complex patterns of metal-vectoring geochemical anomalies in non-linear Earth systems. However, the learning of geochemical background repre-sentations may be suppressed by redundant mutual information from inter-element correlations and by mixed information of elemental concentration data caused by multiplicative cascade geo-processes. To deal with these issues, we conceptualized an idea of a new deep learning architecture called Info-DAN, chaining the information maximization (Infomax) processor to the training network of stacked autoencoders. Infomax is an adaptive learning algorithm from information theory paradigms which aims at maximizing the information flow (joint entropy) passed through a feed-forward neural network processor. It was adopted to encode original multi-element data into independent source signals associated with different geochemical sub-populations and to prevent the dilution of background representations caused by inter-element information redundancy. The recovered source signals were then fed into a DAN processor to assist in modeling the improved representations of geochemical background populations and in enhancing complex anomaly patterns. The Info-DAN technique was applied to stream sediment geochemical data pertaining to the Moalleman district, NE Iran, for performance appraisal in recognition of metal-vectoring geochemical anomalies. Evaluation tools comprising success-rate curves and prediction-area plots indicated that anomaly patterns derived from Info-DAN, compared to those from a stand-alone DAN, reveal a stronger spatial correlation bet

关键词： Geochemical anomaly deep learning Information maximization (Infomax) deep autoencoder network Mineralization

来源：评论

学校读者我要写书评

暂无评论

An end-to-end framework for flight trajectory data analysis based on deep autoencoder network

引用

AEROSPACE SCIENCE AND TECHNOLOGY 2022年第0期127卷

作者： Zhang, Weining Hu, Minghua Du, Jinghan Nanjing Univ Aeronaut & Astronaut Coll Civil Aviat Nanjing Peoples R China

In order to jointly solve the tasks of abnormal trajectory detection and flow pattern recognition for flight trajectory data analysis, an end-to-end framework based on deep autoencoder network is proposed in this paper. Considering the coupling relationship between the two tasks, a structured sparsity-inducing norm is introduced into the reconstruction-based loss function to separate abnormal trajectories from the whole and extract low-dimensional and outliers-free representations from the remaining normal trajectories. On this basis, cluster assignment hardening is applied to further learn cluster-friendly representations as well as cluster assignment for each trajectory. The effectiveness and efficiency of the framework are validated on flight trajectories arriving at Hong Kong International Airport. Experimental results show that the proposed framework not only detects typical spatial anomalies, including holding patterns and rerouting patterns, but also identifies fine-grained cluster structures. Furthermore, it surpasses current state-of-the-art methods in terms of anomaly detection performance and cluster quality, with an improvement of 13.56% and 22.82%, respectively. With parallel computing, its time cost can be reduced to less than 1 second, which helps to perceive traffic situations and monitor abnormal behaviors in real time. (C) 2022 Elsevier Masson SAS. All rights reserved.

关键词： Air traffic Flight trajectory deep autoencoder network Anomaly detection Cluster analysis Meteorological conditions

来源：评论

学校读者我要写书评

暂无评论

Looseness condition feature extraction of viscoelastic sandwich structure using dual-tree complex wavelet packet-based deep autoencoder network

引用

STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL 2020年第3期19卷 873-884页

作者： Si, Yue Zhang, Zhousuo Kong, Chuiqing Li, Shujuan Yang, Guigeng Hu, Bingbing Xian Univ Technol Sch Mech & Instrumental Engn Xian 710048 Shaanxi Peoples R China Xi An Jiao Tong Univ State Key Lab Mfg & Syst Engn Sch Mech Engn Xian Peoples R China China Natl Heavy Machinery Res Inst Co Ltd Xian Peoples R China

It is significant to perform looseness condition detection of viscoelastic sandwich structures to avoid serious accidents. Due to the multilayer characteristic of the viscoelastic sandwich structure, the vibration response signal of such structures is nonlinear and nonstationary. Furthermore, the looseness condition feature signal contained in the vibration response signal is very puny. Condition feature extraction has become a challenging task in the looseness condition detection of viscoelastic sandwich structures. Therefore, a novel method called dual-tree complex wavelet packet-based deep autoencoder network is proposed for this task. First, the vibration response signal of the viscoelastic sandwich structure is decomposed by dual-tree complex wavelet packet transform and the sub-band signals which contain rich energy are extracted. Then, the energies of the extracted sub-band signals are calculated to form a feature set. Finally, a deep autoencoder network is established to fuse the feature set, and the fused feature is viewed as the detection index to detect the looseness condition of the viscoelastic sandwich structure. The proposed method is applied to the connecting bolt looseness condition detection of the viscoelastic sandwich structure to validate its effectiveness. Compared with the detection method based on dual-tree complex wavelet packet transform and energy and the detection method based on dual-tree complex wavelet packet transform and permutation entropy, the results indicate that the effectiveness of the proposed method in this article is more superior to that of the other two methods.

关键词： Bolt looseness condition detection viscoelastic sandwich structure feature extraction dual-tree complex wavelet packet deep autoencoder network

来源：评论

学校读者我要写书评

暂无评论

A Novel Lane-Changing Decision Model for Autonomous Vehicles Based on deep autoencoder network and XGBoost

引用

IEEE ACCESS 2020年 8卷 9846-9863页

作者： Gu, Xinping Han, Yunpeng Yu, Junfu Shandong Univ Key Lab High Efficiency & Clean Mech Manufacture Minist Educ Jinan 250061 Peoples R China Shandong Univ Sch Mech Engn Jinan 250061 Peoples R China

Lane-changing (LC) is a critical task for autonomous driving, especially in complex dynamic environments. Numerous automatic LC algorithms have been proposed. This topic, however, has not been sufficiently addressed in existing on-road manoeuvre decision methods. Therefore, this paper presents a novel LC decision (LCD) model that gives autonomous vehicles the ability to make human-like decisions. This method combines a deep autoencoder (DAE) network with the XGBoost algorithm. First, a DAE is utilized to build a robust multivariate reconstruction model using time series data from multiple sensors;then, the reconstruction error of the DAE trained with normal data is analysed for LC identification (LCI) and training data extraction. Then, to address the multi-parametric and nonlinear problem of the autonomous LC decision-making process, an XGBoost algorithm with Bayesian parameter optimization is adopted. Meanwhile, to fully train our learning model with large-scale datasets, we proposed an online training strategy that updates the model parameters with data batches. The experimental results illustrate that the DAE-based LCI model is able to accurately identify the LC behaviour of vehicles. Furthermore, with the same input features, the proposed XGBoost-based LCD model achieves better performance than other popular approaches. Moreover, a simulation experiment is performed to verify the effectiveness of the decision model.

关键词： Autonomous vehicle lane-changing identification lane-changing decision-making deep autoencoder network XGBoost

来源：评论

学校读者我要写书评

暂无评论

Orthogonal Nonnegative Matrix Factorization using a novel deep autoencoder network

引用

KNOWLEDGE-BASED SYSTEMS 2021年 227卷 107236-107236页

作者： Yang, Mingming Xu, Songhua Xi An Jiao Tong Univ Sch Math & Stat Xian Peoples R China

Orthogonal Nonnegative Matrix Factorization (ONMF) offers an important analytical vehicle for addressing many problems. Encouraged by record-breaking successes attained by neural computing models in solving an assortment of data analytics tasks, a rich collection of neural computing models has been proposed to perform ONMF with compelling performance. Such existing models can be broadly classified into the shallow-layered structure (SLS) based and deep-layered structure (DLS) based models. However, SLS models cannot capture complex relationships and hierarchical information latent in a matrix due to their simple network structures and DLS models rely on an iterative procedure to derive weights, leading to a less efficient solution process and cannot be reused to factorize new matrices. To overcome these shortcomings, this paper proposes a novel deep autoencoder network for ONMF, which is abbreviated as DAutoED-ONMF. Compared with SLS models, the newly proposed model is capable of generating solutions with good interpretability and solution uniqueness like original SLS models, yet the new model attains a superior learning capability thanks to its deep structure employed. In comparison with DLS models, the new model trains a reusable encoder network to directly factorize any given matrix with no need to repeatedly retrain the model for factorizing multiple matrices using a tailor-designed network training procedure. Proof of the procedure's convergence is presented with an analysis of its computational complexity. The numerical experiments conducted on several publicly data sets convincingly demonstrate that the proposed DAutoED-ONMF model gains promising performance in terms of multiple metrics. (C) 2021 Elsevier B.V. All rights reserved.

关键词： Orthogonal Nonnegative Matrix Factorization deep autoencoder network Auxiliary function Multiplication update rule

来源：评论

学校读者我要写书评

暂无评论

Underdetermined blind speech source separation based on deep nearest neighbor clustering algorithm

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2023年第1期82卷 1171-1183页

作者： Niu, Mengdie Zhang, Ye Nanchang Univ Dept Elect Informat Engn Nanchang Jiangxi Peoples R China

In this paper, we propose a new autoencoder network architecture with clustering mechanism for underdetermined blind speech source separation, i.e., the number of mixtures is less than that of sources. The autoencoder network is employed to project the mixtures to embedding space and obtain their embedding vectors. The network model additionally incorporates the clustering mechanism and nearest neighbor clustering algorithm to estimate the clustering centers of the embedding vectors. Then, according to the embedding vectors, the hard and the probability assignment method are proposed to assign the mixtures to their corresponding clusters to recover the sources. The experimental results demonstrate that the proposed method yields better performance compared to the baseline algorithms.

关键词： deep nearest neighbor clustering Underdetermined blind sources separation deep autoencoder network

来源：评论

学校读者我要写书评

暂无评论

Structure and Thrust Fault Detection and Diagnosis for Hypersonic Aircrafts via a Data-Driven Approach 3

Structure and Thrust Fault Detection and Diagnosis for Hyper...

引用

3rd Conference on Fully Actuated System Theory and Applications (FASTA)

作者： Zhang, Chaoyue Hu, Qinglei Han, Tuo Beihang Univ Sch Automat Sci & Elect Engn Beijing 100191 Peoples R China Beihang Univ Sch Aeronaut Sci & Engn Beijing 100191 Peoples R China Beihang Univ Hangzhou Int Innovat Inst Hangzhou 311115 Peoples R China Tianmushan Lab Hangzhou 311115 Peoples R China

ISBN: (纸本)9798350373707;9798350373691

As a complex and significant part of the hypersonic aircraft operating under high speeds and harsh flight conditions, its structure and thrust systems are sometimes fragile to suffer from unforeseeable failures that can seriously degrade the flight performance and even cause mission failures of the aircraft. To deal with such problems from the perspective of online fault detection and diagnosis (FDD) for flight control reconfiguration, a data-driven intelligent FDD scheme for the hypersonic aircraft is proposed in this study considering both the structure and thrust faults. First, a six-degree-of-freedom nonlinear model of the aircraft is employed to generate sampling data of different flight status under various faults. Then, a data preprocessing method and a deep Residual autoencoder network (DRAE) are utilized to predict and classify the structure and thrust faults. To alleviate the vanishing gradient issue, a residual block is integrated in the encoder for efficient training. Numerical simulations show that the FDD algorithm based on the designed scheme can extract the characteristics of different faults for online failure prediction and classification with expected accuracy.

关键词： hypersonic aircraft fault detection and diagnosis deep autoencoder network residual mechanism

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Sentiment-Aware Collaborative Filtering Algorithms for Arabic Recommendation Systems

引用

IEEE ACCESS 2024年 12卷 174441-174454页

作者： Al-Ghuribi, Sumaia Mohammed Noah, Shahrul Azman Mohd Mohammed, Mawal A. Tiwary, Neeraj Saat, Nur Izyan Yasmin Prince Sattam Bin Abdulaziz Univ Coll Comp Engn & Sci Software Engn Dept Al Kharj 11942 Saudi Arabia Taiz Univ Fac Appl Sci Dept Comp Sci Taizz Yemen Univ Kebangsaan Malaysia Fac Informat Sci & Technol Ctr Artificial Intelligent Technol Bangi 43600 Selangor Malaysia

The rapid proliferation of online information necessitates efficient Recommendation Systems (RSs) to assist users in discovering relevant content. While English-language RSs have received significant attention, research on Arabic RSs remains limited despite the increasing demand for Arabic digital content. This paper addresses the scarcity of Arabic-focused Collaborative Filtering (CF) approaches for RS. Recognizing the wealth of information embedded in user reviews, we propose novel review-based CF approaches tailored for Arabic, aiming to enhance recommendation accuracy for Arab users. Our work comprises three key stages: we first develop a comprehensive Arabic lexicon specifically for the book domain. Secondly, using this lexicon we then propose three distinct sentiment-aware ratings, leveraging sentiment analysis of Arabic reviews to enrich traditional rating predictions. Thirdly, these sentiment-aware ratings are integrated into ten diverse CF algorithms from the Surprise library and a deep autoencoder neural network, covering a spectrum of traditional and modern approaches. Extensive experiments conducted on the Large Arabic Book Reviews (LABR) dataset demonstrate the superior performance of our proposed sentiment-aware ratings compared to baseline methods across all evaluated metrics. Further analysis reveals the importance of appropriate sentiment word extraction methods and lexicon selection for accurate sentiment rating calculation. Finally, this study makes a significant contribution to the field of Arabic CF recommendation systems by providing a comprehensive framework for leveraging user review and underscores the importance of further research in this area.

关键词： Reviews Sentiment analysis deep learning Accuracy Prediction algorithms Recommender systems Machine learning algorithms Collaborative filtering Support vector machines Classification algorithms Natural language processing Arabic language sentiment analysis collaborative filtering user reviews deep autoencoder network

来源：评论

学校读者我要写书评

暂无评论

A Tensor Solution for Health Indicator Construction of Metro Wheelset Degradation with Irregular Noise 3

A Tensor Solution for Health Indicator Construction of Metro...

引用

3rd International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2022

作者： Wang, Yu Mao, Wentao Kou, Linlin Liu, Keying School of Computer and Information Engineering Henan Normal University Xinxiang China Beijing Jiaotong University State Key Lab of Rail Traffic Control & Safety Beijing China

ISBN: (纸本)9781665492812

During the operation of metro vehicles, wheel flange and wheel diameter continue to be abrased, so wheel degradation is inevitable. Influenced by the states of welded rail, subgrade and turnout, applied load, speed of running and other components, the degradation process of wheelset is vulnerable to irregular noise interference, which leads to abnormal fluctuation of vibration signal. As a result, the degradation process is hard to be accurately described. To address this problem, this paper proposes a new health indicator construction method for metro wheelset based on tensor reconstruction. First, tensor Tucker decomposition is utilized to obtain the core tensor of original signal, and then tensor reconstruction is applied to transform the signal into a new degradation sequence with noise reduction. Second, the Savitzky-Golay filter is employed to remove the irregular trend from the obtained degradation sequence. Finally, deep autoencoder network is used to extract deep degradation features, and after dimension reduction, a health indicator of wheelset degradation process can be obtained. The effectiveness of the proposed method is verified with Beijing subway wheelset degradation dataset in 2020. The results show that the constructed health indicator can accurately describe the whole degradation process with good trendability and monotonicity. More importantly, the proposed method possess good practicability since the key changing parts in the health indicator can correspond well to the actual maintenance records. © 2022 IEEE.

关键词： deep autoencoder network Health indicator Irregular noise Tucker decomposition

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：