Intelligence Science and Big Data Engineering. Big Data and Machine Learning

Name: Intelligence Science and Big Data Engineering. Big Data and Machine Learning | 9th International Conference, IScIDE 2019, Nanjing, China, October 17-20, 2019, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

9th International Conference, IScIDE 2019, Nanjing, China, October 17-20, 2019, Proceedings, Part II

Zhen Cui Jinshan Pan Shanshan Zhang Liang Xiao Jian Yang(Herausgeber*in)

Springer (Verlag)

Erschienen am 28. November 2019

XX, 455 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-030-36204-1 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Contents - Part II
Contents - Part I
Analysis of WLAN's Receiving Signal Strength Indication for Indoor Positioning
1 Introduction
2 Related Works
3 Theoretical Analysis
4 Measurement Devices and Experimental Environment Settings
5 Characteristic Analysis of Received Signal Strength
5.1 Effect of Antenna Orientation of the Receiver on RSSI
5.2 Effect of Type of Wireless Network Card of the Receiver on RSSI
5.3 Effect of Height Difference of the Transmitting and Receiving Antenna on RSSI
6 Conclusion
References
Computational Decomposition of Style for Controllable and Enhanced Style Transfer
1 Introduction
2 Related Work
3 Methods
3.1 Decomposed by Spectrum Transforms
3.2 Decomposed by Latent Variable Models
3.3 Control Function g
4 Experiments
4.1 Inferiority of Feature Map
4.2 Transfer by Single Style Basis
4.3 Transfer by Intervention
4.4 Transfer by Mixing
4.5 Sketch Style Transfer
4.6 Chinese Painting Style Transfer
5 Conclusion
References
Laplacian Welsch Regularization for Robust Semi-supervised Dictionary Learning
1 Introduction
2 Model Description
2.1 Preliminaries for Semi-supervised Dictionary Learning
2.2 Laplacian Welsch Regularization for Semi-supervised Dictionary Learning
2.3 HQ Optimization of LWR
2.4 Classification Strategy
3 Experiment
3.1 Object Recognition
3.2 Digit Classification
3.3 Convergence Study
4 Conclusion
References
Non-local MMDenseNet with Cross-Band Features for Audio Source Separation
1 Introduction
2 Related Work
2.1 RNN/CNN-Based Approaches
2.2 Generative Approaches
3 Method
3.1 Preliminary
3.2 Non-local MMDenseNet with Cross-Band Features
3.3 Non-local Layer
3.4 Cross-Band Features
4 Experiment
4.1 Dataset
4.2 Metrics and Setting
4.3 Experimental Results
5 Conclusion
References
A New Method of Metaphor Recognition for A-is-B Model in Chinese Sentences
Abstract
1 Introduction
2 Feature Extraction
2.1 Recognition Feature Based on Hyponymy Relation Library
2.2 Recognition Features Based on Sentence Patterns
2.3 Recognition Features Based on Class-Word
2.4 Similarity Feature Based on Word2Vec Calculation
3 Feature-Based Classifier
3.1 SVM Classifier
3.2 Neural Network Classifier
4 Experiment and Evaluation Criteria
4.1 Construction of Metaphor Training Set
4.2 Experiment Platform
4.3 Evaluation Criteria
4.4 Neural Network Training Process
5 Experimental Results and Analysis
6 Conclusion
References
Layerwise Recurrent Autoencoder for Real-World Traffic Flow Forecasting
1 Introduction
2 Related Work
3 Layerwise Recurrent Autoencoder
3.1 Temporal Dependency Modeling
3.2 Spatial Dependency Modeling
4 Experiment
4.1 Experimental Settings
4.2 Experiment Results of Traffic Flow Forecasting
4.3 Experiment Results of Spatial and Temporal Dependency
5 Conclusion
References
Mining Meta-association Rules for Different Types of Traffic Accidents
Abstract
1 Introduction
2 Methodology
2.1 Basic Idea
2.2 Extracting Meta-rules
2.3 Mining Universal Association Rules
3 Experimental Results and Analysis
3.1 Data Preparation
3.2 Parameter Selection
3.3 Rule Generation and Analysis
4 Conclusions
Acknowledgements
References
Reliable Domain Adaptation with Classifiers Competition
1 Introduction
2 Related Work
3 Reliable Domain Adaptation
3.1 Problem Definition
3.2 Model Formulation
4 Experiment
4.1 Data Preparation
4.2 Experimental Setting
4.3 Experimental Results
4.4 Model Analysis and Discussion
5 Conclusion
References
An End-to-End LSTM-MDN Network for Projectile Trajectory Prediction
Abstract
1 Introduction
2 Related Work
3 LSTM-MDN Prediction Network
3.1 LSTM
3.2 MDN
3.3 Fully Connected Network
3.4 Loss Function
4 Experiment Results and Analysis
4.1 Dataset
4.2 Network and Training
4.3 Performance Metric
4.4 Results and Analysis
5 Conclusion
References
DeepTF: Accurate Prediction of Transcription Factor Binding Sites by Combining Multi-scale Convolution and Long Short-Term Memory Neural Network
1 Introduction
2 Materials and Methods
2.1 Benchmark Datasets
2.2 Feature Representation
2.3 Model Architecture
2.4 Evaluation Metric
3 Results
3.1 Comparisons Among CNN-LSTM, CNN, and LSTM
3.2 MNOH Helps to Improve the Prediction Performance
3.3 Multi-scale Convolution Outperforms in Prediction
3.4 Comparisons with Existing Predictors
4 Conclusions
References
Epileptic Seizure Prediction Based on Convolutional Recurrent Neural Network with Multi-Timescale
Abstract
1 Introduction
2 Proposed Model
2.1 Data Segmentation
2.2 Feature Extraction
2.3 Multi-Timescale Convolutional Recurrent Neural Network
3 Dataset
4 Experiments and Results
4.1 Evaluation Metrics
4.2 Results
5 Conclusion
Acknowledgements
References
L2R-QA: An Open-Domain Question Answering Framework
1 Introduction
2 Related Work
2.1 Question Answering Systems (QASs)
2.2 QA with Wikipedia
3 Datasets
3.1 Wikipedia
3.2 SQuAD
4 Framework
4.1 Information Retrieval and Topic Model
4.2 Paragraph Features Extraction
4.3 Question Semantic Modeling
4.4 Candidate Answers Selection
5 Experiments
5.1 Document Retrieval Based on LSI
5.2 Document Reader
5.3 Open-Domain Question Answering System
6 Conclusion
References
Attention Relational Network for Few-Shot Learning
1 Introduction
2 Related Work
2.1 Relational Network
2.2 Attention Mechanism
3 Attention Relational Network
3.1 Problem Description
3.2 Model of Attention Relational Network
3.3 Network Module and Optimization
4 Experiment
4.1 Experiment on MiniImageNet Dataset
4.2 Experiment on Omniglot Dataset
4.3 Analysis
5 Conclusion
References
Syntactic Analysis of Power Grid Emergency Pre-plans Based on Transfer Learning
Abstract
1 Introduction
1.1 Background
1.2 Related Work
2 Syntax Analysis Method of the Emergency Pre-plans Based on Transfer Learning
2.1 Model Introduction
2.1.1 ULMFiT Model
2.1.2 Modified ULMFiT Model
2.2 Training Methods Based on Transfer Learning
2.2.1 Word Vector Training Method of the Emergency Pre-plans Based on Transfer Learning
2.2.2 Training Method of Syntax Analysis Model Based on Transfer Learning
2.3 Syntactic Analysis
2.3.1 Syntactic Analysis
2.3.2 Sample Results
3 Experiment and Results
3.1 Experiment Setting
3.2 Experimental Results
4 Conclusion
References
Improved CTC-Attention Based End-to-End Speech Recognition on Air Traffic Control
Abstract
1 Introduction
2 Original CTC-Attention Encoder-Decoder
2.1 Connectionist Temporal Classification (CTC)
2.2 Attention Based Encoder-Decoder
2.3 Original CTC-Attention Encoder-Decoder
3 Improved CTC-Attention Encoder-Decoder
3.1 VggCNN and Bidirectional LSTM Based Encoder
3.2 CTC-Attention Based Decoder
4 Experiments
4.1 ATC Corpus
4.2 Training Tricks
4.3 Experimental Setup
4.4 LSTM Based Language Model Using Character Units
5 Result
6 Conclusion
Acknowledgements
References
Revisit Lmser from a Deep Learning Perspective
1 Introduction
2 Related Work
2.1 Networks with Symmetrically Weighted Connections
2.2 Networks with Symmetrically Skip Connections
3 A Brief Review of Lmser
4 Methods
4.1 Revisit Lmser and Implement It on Multiple Fully-Connected Layers
4.2 Jointly Supervised and Unsupervised Lmser Learning
5 Experiments
5.1 Datasets and Experimental Settings
5.2 Reconstruction
5.3 Recognition
5.4 Generation
5.5 Association
6 Conclusion
References
A New Network Traffic Identification Base on Deep Factorization Machine
Abstract
1 Introduce
2 Method
3 Experiment
3.1 DataSet and Processing of Missing Values and Imbalanced Data
3.2 Metrics
3.3 Experimental Results and Analysis
4 Conclusion
References
3Q: A 3-Layer Semantic Analysis Model for Question Suite Reduction
1 Introduction
2 Related Work
2.1 QG and QA
3 Question Analysis Model
3.1 First Layer Analysis
3.2 Second Layer Analysis
3.3 Third Layer Analysis
4 Downsize of Questions Set
4.1 Screening Strategy
5 Evaluation
5.1 Experiment Proof
5.2 Computation of Information Entropy
5.3 Grade Level
6 Experiment
6.1 Data and Model
6.2 Comparative Methods
6.3 Result
7 Conclusion
References
Data Augmentation for Deep Learning of Judgment Documents
1 Introduction
2 Related Work
2.1 Legal Judgement Prediction
2.2 Data Augmentation
3 Experiments
3.1 Datasets and Preprocessing
3.2 Data and Augmentation
3.3 FastText
3.4 TextCNN
4 Results and Discussion
5 Conclusion
References
An Advanced Least Squares Twin Multi-class Classification Support Vector Machine for Few-Shot Classification
Abstract
1 Introduction
2 Related Work
3 Advanced Least Squares Twin Multi-class Classification Support Vector Machine
3.1 Linear ALST-KSVC
3.2 Kernel ALST-KSVC
3.3 Decision Function
4 Experiments
4.1 Datasets
4.2 Experimental Results
5 Conclusion
References
LLN-SLAM: A Lightweight Learning Network Semantic SLAM
Abstract
1 Introduction
2 Semantic SLAM
2.1 Related Work
2.2 Semantic Extraction
3 LLN-SLAM and Global Semantic Database Update
3.1 LLN-SLAM
3.2 Semantic Map Storage Structure Design
3.3 Global Semantic Database Update
4 Experiment and Result Analysis
4.1 LNN SLAM Semantic Accuracy Test
4.2 LNN SLAM Positioning Accuracy Test
5 Conclusion
References
Meta-cluster Based Consensus Clustering with Local Weighting and Random Walking
1 Introduction
2 Proposed Approach
2.1 Formulation
2.2 Cluster-Wise Similarity Graph
2.3 Random Walk Propagation on Clusters
2.4 Consensus by Locally Weighted Majority Voting
3 Experiments
3.1 Datasets and Experimental Setting
3.2 Comparison with Other Consensus Clustering Methods
3.3 Robustness to Ensemble Size M
3.4 Sensitivity of Parameter T
3.5 Time Cost
4 Conclusion
References
Robust Nonnegative Matrix Factorization Based on Cosine Similarity Induced Metric
1 Introduction
2 Related Work
2.1 NMF
2.2 CIMNMF
2.3 RSNMF
3 The Proposed CSNMF Approach
3.1 Cosine Similarity Induced Metric
3.2 The CSNMF Model
3.3 Theoretical Analysis on CSNMF
4 Experimental Results
4.1 Face Image Databases
4.2 Results on FERET Database
4.3 Results on Yale B Database
4.4 Convergence Verification
5 Conclusion
References
Intellectual Property in Colombian Museums: An Application of Machine Learning
Abstract
1 Introduction
2 Context
2.1 Museum Networks in Colombia
2.2 Recent Studies on Intellectual Property
2.3 Application of Machine Learning in Museums
3 Method
3.1 Data
3.2 Variables Used
3.3 Algorithms Used
4 Results
5 Conclusions
References
Hybrid Matrix Factorization for Multi-view Clustering
1 Introduction
2 Overview of NMF and SymNMF
2.1 NMF
2.2 SymNMF
3 HMF
3.1 Formulation
3.2 Optimization
4 Experiments
4.1 Experimental Setup
4.2 Results
4.3 Convergence
5 Conclusion
References
Car Sales Prediction Using Gated Recurrent Units Neural Networks with Reinforcement Learning
Abstract
1 Introduction
2 Related Work
3 Data Analysis and the Proposed Method
3.1 Analysis of Car Sales Data Set
3.2 Feature Selection
3.3 The Proposed GRURL Model
4 Experiments
4.1 Experimental Settings
4.2 Experimental Results
5 Conclusion
Acknowledgment
References
A Multilayer Sparse Representation of Dynamic Brain Functional Network Based on Hypergraph Theory for ADHD Classification
Abstract
1 Introduction
2 Materials and Methods
2.1 Material Preparation
2.2 Construction of Hyper Network Based on the Hyper Graph Theory
2.3 Sparse Feature Extraction
3 Experimental Process and Results Analysis
4 Conclusion
References
Stress Wave Tomography of Wood Internal Defects Based on Deep Learning and Contour Constraint Under Sparse Sampling
1 Introduction
2 Proposed Method
2.1 Stress Wave Tomography Using Deep Learning and Contour Constraint
2.2 Data Acquisition
3 Experiment Results and Analysis
3.1 Stress Wave Tomography Results Under Sparse Sampling
3.2 Effect of Object Detection Results on Tomography
4 Conclusion
References
Robustness of Network Controllability Against Cascading Failure
Abstract
1 Introduction
2 Model
2.1 Controllability of Complex Networks
2.2 Cascading Failure Load-Capacity Model
2.3 Attack Strategies
2.4 Robustness Metric for Network Controllability
3 Simulation Experiments and Results
3.1 Random Attack Experiments
3.2 Malicious Attack Experiments
4 Conclusion
References
Multi-modality Low-Rank Learning Fused First-Order and Second-Order Information for Computer-Aided Diagnosis of Schizophrenia
Abstract
1 Introduction
2 Background
2.1 First-Order and Second-Order Brain Functional Connection Network
2.2 Low-Rank Representation
2.3 Materials
3 Method
3.1 Proposed Method
3.2 Optimization and Solution
4 Experiments and Discussion
4.1 Comparison of Our Method and Baseline Classification Methods
4.2 Comparison of Our Method and State of Art Multi-modality Based Methods
4.3 Analysis of Convergence and Parameter Sensitivity
5 Conclusion
Acknowledgments
References
A Joint Bitrate and Buffer Control Scheme for Low-Latency Live Streaming
1 Introduction
2 System Overview
3 Joint Bitrate and Buffer Control Scheme
3.1 Inputs
3.2 Outputs
3.3 Actor-Critic Network
3.4 QoE Metrics
3.5 Training
4 Experiments and Results
5 Conclusion
References
Causal Discovery of Linear Non-Gaussian Acyclic Model with Small Samples
1 Introduction
2 Related Work
3 Background
3.1 LiNGAM and Preliminaries
3.2 Effects of Small Samples on the LiNGAM
4 MiS-LiNGAM: Minimal Set-Based LiNGAM Algorithm
5 Experiments and Discussion
5.1 Results on Simulated Networks
5.2 Results on Real-World Networks
6 Conclusion
References
Accelerate Black-Box Attack with White-Box Prior Knowledge
1 Introduction
2 Related Work
3 Multi-model Query Black-Box Attack
3.1 Black-Box Attack Loss Function and Optimization
4 Experimental Evaluation
4.1 Setup
4.2 Attack Performance
5 Conclusion
References
A Dynamic Model + BFR Algorithm for Streaming Data Sorting
1 Introduction
2 A Dynamic Model for Streaming Data Sorting
2.1 Step 1: Find the Range of Streaming Data
2.2 Step 2: Fit the Distribution of Streaming Data
2.3 Step 3: Estimate the Sorting Result of Current Data
3 The Dynamic Model + BFR Algorithm
3.1 The BFR Algorithm
3.2 Combine BFR Algorithm with the Dynamic Model
4 Experimental Study
4.1 Experiments Based on the Dynamic Model
4.2 Experiments Based on Dynamic Model+BFR Algorithm
5 Conclusion and Future Work
References
Smartphone Behavior Based Electronical Scale Validity Assessment Framework
Abstract
1 Introduction
2 Related Work
3 Methodology
3.1 Framework Design
3.2 WeChat Application
3.3 Backend
3.4 Inventory
4 Validity Assessment
4.1 Experiment
4.2 Feature Extraction and Classification
4.3 Result
5 Discussion
Acknowledgment
References
Discrimination Model of QAR High-Severity Events Using Machine Learning
1 Introduction
2 Related Works
3 Methodology
3.1 Data Resampling
3.2 Dimensionality Reduction and Feature Elimination
3.3 Discrimination Method Based on RF
4 Experimental Results
4.1 Evaluation Indicators
4.2 QHSE Dataset
4.3 Results Overview
4.4 Examples of Anomalous Flights
5 Conclusion
References
A New Method of Improving BERT for Text Classification
1 Introduction
2 Related Work
2.1 Deep Neural Networks
2.2 Pre-training Model
3 BERT-CNN Model
3.1 Lexicon Encoder
3.2 Multi-layer Transformer Encoder
3.3 Local CNN Encoder
3.4 Transformer Encoder
3.5 Output Layer
4 Experiments
4.1 Data Sets
4.2 Experiments Setup
4.3 Baselines and Result
4.4 Further Analysis
5 Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken