检索结果-内蒙古大学图书馆

A deep learning method based on convolutional neural network for automatic modulation classification of wireless signals

引用

WIRELESS NETWORKS 2019年第7期25卷 3735-3746页

作者： Xu, Yu Li, Dezhi Wang, Zhenyong Guo, Qing Xiang, Wei Harbin Inst Technol Sch Elect & Informat Engn Harbin Peoples R China

Automatic modulation classification plays an important role in many fields to identify the modulation type of wireless signals in order to recover signals by demodulation. In this paper, we contribute to explore the suitable architecture of deep learning method in the domain of communication signal recognition. Based on architecture analysis of the convolutional neural network, we used real signal data generated by instrument as dataset, and achieved compatible recognition accuracy of modulation classification compared with several representative structure. We state that the deeper network architecture is not suitable for the signal recognition due to its different characteristic. In addition, we also discuss the difficult of training algorithm in deep learning methods and employ the transfer learning method in order to reap the benefits, which stabilize the training process and lift the performance. Finally, we adopt the denoising autoencoder to preprocess the received data and provide the ability to resist finite perturbations of the input. It contributes to a higher recognition accuracy and it also provide a new idea to design the denoising modulation recognition model.

关键词： Modulation classification Deep learning Convolutional neural network Wireless signal Transfer learning Denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

DGHNL: A new deep genetic hierarchical network of learners for prediction of credit scoring

引用

INFORMATION SCIENCES 2020年第0期516卷 401-418页

作者： Plawiak, Pawel Abdar, Moloud Plawiak, Joanna Makarenkov, Vladimir Acharya, U. Rajendra Polish Acad Sci Inst Theoret & Appl Informat Baltycka 5 PL-44100 Gliwice Poland Cracow Univ Technol Fac Comp Sci & Telecommun Dept Informat & Commun Technol Warsaw 24 StF-3 PL-31155 Krakow Poland Univ Quebec Montreal Dept Comp Sci Montreal PQ H2X 3Y7 Canada AGH Univ Sci & Technol Dept Biocybernet & Biomed Engn Fac Elect Engn Automat Comp Sci & Biomed Engn PL-30059 Krakow Poland Ngee Ann Polytech Dept Elect & Comp Engn Clementi 599491 Singapore SUSS Univ Sch Sci & Technol Dept Biomed Engn Clementi 599491 Singapore Kumamoto Univ IROAST Kumamoto Japan

Credit scoring (CS) is an effective and crucial approach used for risk management in banks and other financial institutions. It provides appropriate guidance on granting loans and reduces risks in the financial area. Hence, companies and banks are trying to use novel automated solutions to deal with CS challenge to protect their own finances and customers. Nowadays, different machine learning (ML) and data mining (DM) algorithms have been used to improve various aspects of CS prediction. In this paper, we introduce a novel methodology, named Deep Genetic Hierarchical Network of Learners (DGHNL). The proposed methodology comprises different types of learners, including Support Vector machines (SVM), k-Nearest Neighbors (kNN), Probabilistic Neural Networks (PNN), and fuzzy systems. The Statlog German (1000 instances) credit approval dataset available in the UCI machine learning repository is used to test the effectiveness of our model in the CS domain. Our DGHNL model encompasses five kinds of learners, two kinds of data normalization procedures, two extraction of features methods, three kinds of kernel functions, and three kinds of parameter optimizations. Furthermore, the model applies deep learning, ensemble learning, supervised training, layered learning, genetic selection of features (attributes), genetic optimization of learners parameters, and novel genetic layered training (selection of learners) approaches used along with the cross-validation (CV) train-ingtesting method (stratified 10-fold). The novelty of our approach relies on a proper flow and fusion of information (DGHNL structure and its optimization). We show that the proposed DGHNL model with a 29-layer structure is capable to achieve the prediction accuracy of 94.60% (54 errors per 1000 classifications) for the Statlog German credit approval data. It is the best prediction performance for this well-known credit scoring dataset, compared to the existing work in the field. (C) 2019 The Authors. Publish

关键词： Credit scoring machine learning Data mining Ensemble learning Deep learning Genetic algorithm Feature extraction and selection

来源：评论

学校读者我要写书评

暂无评论

Investigation of Feature Inputs for Binary Classification of Ultrasonic ndT signals using SVM and Neural Networks 62

Investigation of Feature Inputs for Binary Classification of...

引用

62nd IEEE international Midwest Symposium on Circuits and Systems (MWSCAS)

作者： Virupakshappa, Kushal Oruklu, Erdal IIT Dept Elect & Comp Engn Chicago IL 60616 USA

ISBN: (纸本)9781728127880

Classification of Ultrasonic Non-Destructive Testing (ndT) signals can be done by machine learning models including Support Vector machine (SVM) and Neural Networks (NN). The main objective of this study is to classify the ultrasonic A-scan data either as flaw echoes or clutter echoes (no flaw). The signal pre-processing has been done using Discrete Wavelet Transform (DWT). The refined low pass output has been used as feature input to the machine learning algorithms either directly or as a power signal. In case of SVM, direct low pass output in windowed format was tested with linear kernel and Radial Basis Function (RBF) kernel and the power signal of the low pass DWT in the windowed format was also tested with linear kernel and RBF kernel. SVM simulation results show that the direct low pass signal with linear kernel fails to converge while power of the low pass DWT achieves an accuracy of around 95%. RBF kernel accuracy was around 98% irrespective of the format of the signal. In case of Neural Network, both the direct low pass output and the power of the low pass output were tested and it was found that the direct low pass output with NN yielded 94% accuracy while the power of the low pass output with NN yielded an accuracy of 98%.

关键词： Ultrasound ndT machine learning Neural Networks Support Vector machine Unconstrained Optimization Constrained Optimization

来源：评论

学校读者我要写书评

暂无评论

On Constrained Nonconvex Stochastic Optimization: A Case Study for Generalized Eigenvalue Decomposition 22

On Constrained Nonconvex Stochastic Optimization: A Case Stu...

引用

22nd international conference on Artificial Intelligence and Statistics (AISTATS)

作者： Chen, Zhehui Li, Xingguo Yang, Lin Haupt, Jarvis Zhao, Tuo Georgia Tech Atlanta GA 30332 USA Princeton Univ Princeton NJ 08544 USA Univ Minnesota Minneapolis MN 55455 USA

We study constrained nonconvex optimization problems in machine learning and signal processing. It is well-known that these problems can be rewritten to a min-max problem in a Lagrangian form. However, due to the lack of convexity, their landscape is not well understood and how to find the stable equilibria of the Lagrangian function is still unknown. To bridge the gap, we study the landscape of the Lagrangian function. Further, we define a special class of Lagrangian functions. They enjoy the following two properties: *** are either stable or unstable (Formal definition in Section 2);*** equilibria correspond to the global optima of the original problem. We show that a generalized eigenvalue (GEV) problem, including canonical correlation analysis and other problems as special examples, belongs to the class. Specifically, we characterize its stable and unstable equilibria by leveraging an invariant group and symmetric property (more details in Section 3). Motivated by these neat geometric structures, we propose a simple, efficient, and stochastic primal-dual algorithm solving the online GEV problem. Theoretically, under sufficient conditions, we establish an asymptotic rate of convergence and obtain the first sample complexity result for the online GEV problem by diffusion approximations, which are widely used in applied probability. Numerical results are also provided to support our theory.

关键词： signal processing

来源：评论

学校读者我要写书评

暂无评论

Development of the Virtual Museum of Nonthaburi 19

Development of the Virtual Museum of Nonthaburi

引用

2nd international conference on Image and Graphics processing (ICIGP) / 5th international conference on Virtual Reality

作者： Mingsiritham, Kemmanat Chanyawudhiwan, Gan Sukhothai Thammathirat Open Univ Off Educ Technol Nonthaburi Thailand

ISBN: (纸本)9781450360920

Nonthaburi is a city for learning with history, local wisdoms, art and culture, and lifestyle. It is necessary to establish a learning resource to create the feeling of bonding and protection of the city. However, the problem is the visitor needs to travel to the learning resource. The use of technology enables an easy and convenient access to the learning resource, leading to preservation of the culture. This research has an objective to develop the Virtual Museum of Nonthaburi by interviewing with curators and/or staff of Nonthaburi Museum and experts in virtual learning resources. It studies the using result of the Virtual Museum of Nonthaburi from general visitors. Data was analyzed by using mean, standard deviation and content analysis. Research results: 1) The model of the Virtual Museum of Nonthaburi consists of 6 components: 1) information, 2) media and tools used, 3) interaction, 4) design, 5) decision support system, and 6) supporting factors. The overall quality result found that the quality was at the highest level )(X) over bar = 4.51, S.D =. 0.57( and 2) Using results of the Virtual Museum of Nonthaburi found that most visitors viewed that Nonthaburi had a lot of tourist attractions with historical significance, cultural value, and living aspect. Especially, pottery which presents the development from the past to present is beautiful local artwork that should be preserved. The overall satisfaction result found that the quality was at the high level ((X) over bar = 4.42, S.D = .0.65).

关键词： Virtual Museum Virtual Reality Virtual learning resources

来源：评论

学校读者我要写书评

暂无评论

NLP chatbot for Discharge Summaries 2

NLP chatbot for Discharge Summaries

引用

2nd international conference on Intelligent Communication and Computational Techniques (ICCT)

作者： Lal, Harsh Lal, Priyanshu Univ Calif San Diego La Jolla CA 92093 USA Birla Inst Technol Hyderabad Telangana India

ISBN: (纸本)9781728117119

One of the most important goals in health-care is improving the quality of health and service rendered by healthcare industry. There has been a growing demand for ease of access to health-care, and this necessitates the development of health-care technology with Human Computer Interaction(HCI) design in mind. HCI usage can help both patients and healthcare providers as well[3]. With a wealth of information available to patients and providers, it can be difficult to find relevant information in a timely manner, especially for providers with busy schedules and patients with limited health literacy. We propose a Human Computer Interaction system that can answer patient ' s question and help improve the quality of health-care services provided to an individual. We make use of Natural Language processing, machine learning and Big Data technologies to provide a feasible deploy-able solution. for the same.

关键词： Natural Language processing Chatbot Human Computer Interaction machine learning Big Data

来源：评论

学校读者我要写书评

暂无评论

An Optimized Iterative Partitioning Model for Predicting Computer System Failures Based on Deep learning 1

引用

2nd international conference on Advanced Intelligent Systems for Sustainable Development, AI2SD 2019

作者： El Fouki, Mohammed Aknin, Noura El Kadiri, Kamal Eddine TIMS Research Unit LIROSA Laboratory Abdelmalek Essaadi University Tetouan Morocco

ISBN: (数字)9783030366742

ISBN: (纸本)9783030366735

Computer systems are becoming increasingly complex, they are growing dynamically due to the mobility of their devices, multi-operating environments, frequent updates and upgrades. Conventional reliability theory and conventional methods rarely take into account the actual state of the system;thus, they are not able to reflect the dynamics of the execution systems and the failure processes. The variety of models used illustrate the lack of consensus on how to apprehend this phenomenon, especially the difficulties to formulate a general model of interpretation. Moreover, this diversity of approaches has been developed mainly to diagnosis computer systems. Consequently, the nature of the information available on these systems and the type of defects to be detected lead to the implementation of specific strategies. The main objective of this paper is to use an Iterative Partitioning Log Mining algorithm (IPLOM) to cluster logs events before they are processed further by an optimized deep learning model to detect anomalies form large amount of IBM BlueGene/L collected data. ALL the features are classified by the score and also selected to be saved or even removed from the dataset. The procedures are frequently univariate and consider the feature independently, or with regard to the dependent variable. In such methods, the subset selection process is independent of the learning algorithm, and it is a pre-processing step. Generally, this work describes how to optimize deep learning models in different types of machine learning problems. We started by developing a mining model, and then apply a filter to the input data in order to eliminate the correlated information. Then, we compared the results of these different models using an elevation curves graph. Finally, we extracted additional knowledge from the underlying mining structure. This leads to optimize learning features and to reduce the classification time and even improve the performance of systems’ rules, particular

关键词： Data mining

来源：评论

学校读者我要写书评

暂无评论

learning beamforming in ultrasound imaging 2

Learning beamforming in ultrasound imaging

引用

2nd international conference on Medical Imaging with Deep learning (MIDL)

作者： Vedula, Sanketh Senouf, Ortal Zurakhov, Grigoriy Bronstein, Alex Michailovich, Oleg Zibulevsky, Michael Technion Haifa Israel Univ Waterloo Waterloo ON Canada

Medical ultrasound (US) is a widespread imaging modality owing its popularity to cost efficiency, portability, speed, and lack of harmful ionizing radiation. In this paper, we demonstrate that replacing the traditional ultrasound processing pipeline with a data-driven, learnable counterpart leads to significant improvement in image quality. Moreover, we demonstrate that greater improvement can be achieved through a learning-based design of the transmitted beam patterns simultaneously with learning an image reconstruction pipeline. We evaluate our method on an in-vivo first-harmonic cardiac ultrasound dataset acquired from volunteers and demonstrate the significance of the learned pipeline and transmit beam patterns on the image quality when compared to standard transmit and receive beamformers used in high frame-rate US imaging. We believe that the presented methodology provides a fundamentally different perspective on the classical problem of ultrasound beam pattern design.

关键词： Ultrasound Imaging Deep learning Beamforming

来源：评论

学校读者我要写书评

暂无评论

Generative Object Detection: Erasing the Boundary via Adversarial learning with Mask 2

Generative Object Detection: Erasing the Boundary via Advers...

引用

2nd IEEE international conference on Information Communication and signal processing (ICICSP)

作者： Jang, Heeoh Kim, Dongkyu Ahn, Wonhyuk Lee, Heung-Kyu Korea Adv Inst Sci & Technol Sch Comp Daejeon South Korea Samsung Elect Suwon South Korea

ISBN: (纸本)9781728151021

Recently, object detection has presented superior performance using the deep convolutional neural network (CNN). However, most CNN-based models need the bounding box information of the input image in pairs. To overcome this limitation, we propose the Generative Object Detection which learns with only cropped images that are not in pairs. Our model based on Generative Adversarial Networks (GAN) creates cropped images by making a mask that represents the object region. To achieve this goal, we devise a novel mask mean loss (MML) that helps the GAN be able to estimate the distribution of training data and uses dilated convolution for a wider reception field in the generator. The experimental results show that Generative Object Detection improves the mIoU and accuracy.

关键词： generative adversarial network object detection deep-learning

来源：评论

学校读者我要写书评

暂无评论

Deep Multimodal Network: a multi-signal input neural network for Ad Click Predictions 19

Deep Multimodal Network: a multi-signal input neural network...

引用

2nd international conference on Big Data Technologies (ICBDT) / 3rd international conference on Business Information Systems Workshop (ICBIS)

作者： Chen, Yang Du, Guanglei Wang, Xu Natl Univ Def Technol Sci & Technol Parallel & Distributed Lab Changsha Peoples R China

ISBN: (纸本)9781450371926

Online advertisement is an important source of revenue for internet companies, so increasing click-through rates (CTR) on ads is crucial. the traditional CTR prediction model only extracts the classification features and numerical features of the advertisement, but ignores the title, description and other text features of the advertisement. However, we know that such information is very important for an advertisement. Therefore, we propose a deep multimodal network (DMN) to solve this problem. on the basis of the traditional deep model, DMN add the text features of cyclic neural network learning, so as to improve the performance of the model. we also did a lot of visual data analysis on the dataset, Finally, we conducted a comparative experiment on the real-world dataset.

关键词： Multimodal Deep learning CTR

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：