PRICAI 2018: Trends in Artificial Intelligence

Name: PRICAI 2018: Trends in Artificial Intelligence | 15th Pacific Rim International Conference on Artificial Intelligence, Nanjing, China, August 28-31, 2018, Proceedings, Part II
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

15th Pacific Rim International Conference on Artificial Intelligence, Nanjing, China, August 28-31, 2018, Proceedings, Part II

Xin Geng Byeong-Ho Kang(Herausgeber*in)

Springer (Verlag)

Erschienen am 30. Juli 2018

XXIX, 524 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-319-97310-4 (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
An Improved Artificial Immune System Model for Link Prediction
Abstract
1 Introduction
2 Related Work
3 Improved Artificial Immune System Model for Link Prediction
3.1 Definition of Dynamic Link Prediction Features
3.2 Algorithm for Link Prediction
4 Experimental Evaluation
4.1 Dataset
4.2 Performance Evaluation with Different Baseline Methods in Literature
4.3 Analysis of Different Factors' Impacts in Link Prediction
5 Conclusion
References
Anchored Projection Based Capped l2,1-Norm Regression for Super-Resolution
1 Introduction
2 Related Work
2.1 Simple Functions for Super-Resolution
2.2 Anchored Neighborhood Regression (ANR)
3 Proposed Method
3.1 Capped l2,1-Norm Regression
3.2 Converge Analysis
4 Experimental Results
4.1 Experimental Settings
4.2 Parameters Analysis
4.3 Compared Methods
5 Conclusion
References
Query Expansion Based on Semantic Related Network
Abstract
1 Introduction
2 Related Work
3 Problem Description and Definitions
4 Semantic Related Network Construction
5 Heuristic Query Expansion
6 Experimental Results
6.1 Accuracy
6.2 Effectiveness
7 Conclusion
Acknowledgement
References
Improving the Stability for Spiking Neural Networks Using Anti-noise Learning Rule
Abstract
1 Introduction
2 The SRM Neuron Model and Learning Rule
3 The Anti-noise SNN Learning Rule
4 Experimental Results
4.1 XOR Task
4.2 WBC Task
5 Conclusions
Acknowledgement
References
An Improved Convolutional Neural Network Model with Adversarial Net for Multi-label Image Classification
1 Introduction
2 Improved CNN Model
3 Using Adversarial Network to Improve Accuracy
3.1 Adversarial Spatial Dropout Network Training
3.2 Joint Learning
4 Experiment
4.1 Datasets and Evaluation Measures
4.2 Image-Fine-Tuning on Multi-label Image Set
4.3 Results on Corel 5K and VOC 2012
5 Conclusion
References
Integrating Multiscale Contrast Regions for Saliency Detection
Abstract
1 Introduction
2 Related Work
3 Integrating Multiscale Contrast Regions Deep Network
4 Multi-level Maps Fusion
5 Experiment Result
5.1 Performance Comparison
5.2 Speed Improvement
6 Conclusions
References
Automatic Conditional Generation of Personalized Social Media Short Texts
Abstract
1 Introduction
2 The Conditional Language Generation Model
2.1 Psychological Text Classification Model
2.2 The Text Generator
3 Text Generation Results
3.1 Newly-Generated Text Samples
3.2 The Human Evaluation of Generated Texts
3.3 The BFP Scores of Generated Texts
4 Conclusions and Limitations
Acknowledgement
References
Deep Multi-modal Learning with Cascade Consensus
1 Introduction
2 Related Work
3 Proposed Method
3.1 Deep Canonical Correlation Analysis (DCCA)
3.2 Cascade Deep Multi-Modal Networks (CDMM)
4 Experiments
4.1 Datasets and Configurations
4.2 Comparing with CCA-Based Multi-modal Methods
4.3 Investigation on Embedding of Different Layers
4.4 Empirical Investigation on Convergence
5 Conclusion
References
Driving the Narrative Flow of an Interactive Storytelling System for Case Studies
Abstract
1 Introduction
2 Storytelling Knowledge
2.1 Event Model
2.2 Domain Knowledge Base
2.3 Story World Model
3 Planning the Case Narrative
4 Results and Analysis
5 Conclusion
References
Pum-Riang Thai Silk Pattern Classification Using Texture Analysis
1 Introduction
2 Previous Work
3 Design
3.1 Image Enhancement
3.2 Feature Extraction
3.3 Feature Selection
3.4 Classifier Training
4 Evaluation
4.1 Methodology
4.2 Feature Extraction
4.3 Feature Selection
4.4 Number of Features
4.5 Classifier
4.6 Prediction Performance on Test Set
5 Conclusion
References
Fuzzy Rough Based Feature Selection by Using Random Sampling
Abstract
1 Introduction
2 Preliminaries
2.1 FRS
2.2 Attribute Reduction Based on FRS
3 Random Sampling Based FRS
4 Random Reduction Algorithm
5 Numerical Experiment
5.1 Compare CAR with RAR
5.2 Compare IAR with RAR
6 Conclusions
References
Segmenting Sound Waves to Support Phonocardiogram Analysis: The PCGseg Approach
1 Introduction
2 Previous Work
2.1 Segmenting Point Series
2.2 PCG Segmentation
3 Formalism
4 PCGseg
4.1 Motif Detection
5 Evaluation
5.1 Evaluation Data Set
5.2 Runtime Evaluation
5.3 Classification Accuracy
6 Discussion
7 Conclusions
References
A Lazy One-Dependence Classification Algorithm Based on Selective Patterns
1 Introduction
2 Background
3 A Lazy Classification Algorithm Based on Selective Patterns
3.1 Characterization of Discriminative Patterns
3.2 A Lazy One-Dependence Classification Algorithm
4 Experiments and Evaluations
4.1 Parameter Analysis
4.2 Empirical Setup
4.3 Error Rate Analysis
5 Conclusion and Future Work
References
A Client-Assisted Approach Based on User Collaboration for Indoor Positioning
Abstract
1 Introduction
2 Proposed Approach
2.1 CA System Model
2.2 CA Algorithm
3 Influence of Density Distribution
3.1 Effects of AP Density Distribution
3.2 The Influence of User Distribution
4 Implementation and Evaluation
4.1 Environmental Setup
4.2 Efficiency Improvement
5 Conclusion and Future Work
References
Achieving Multiagent Coordination Through CALA-rFMQ Learning in Continuous Action Space
1 Introduction
2 Preliminaries
2.1 CALA
2.2 rFMQ
3 CALA-rFMQ
3.1 Optimum Discrete Actions Using rFMQ with PHC
3.2 Win or Learn Slow Continuous Action Learning Automata (WoLS-CALA)
4 Experimental and Conclusion
References
Environmental Reconstruction for Autonomous Vehicle Based on Image Feature Matching Constraint and Score
Abstract
1 Introduction
2 Proposed Framework
3 Experiment
4 Conclusion
References
An Improved Particle Filter Target Tracking Algorithm Based on Color Histogram and Convolutional Network
1 Introduction
2 Framework of BCH-CN-PF Algorithm
3 Description of BCH-CN-PF Algorithm
3.1 Image Preprocessing
3.2 Block Color Histogram
3.3 Convolutional Network Model
3.4 Improved Particle Filtering Algorithm
4 Experiments and Analysis
4.1 Quantitative Analysis
4.2 Qualitative Analysis
5 Summary
References
Mini-Batch Variational Inference for Time-Aware Topic Modeling
1 Introduction
2 Method
3 Experiment
4 Related Work
5 Conclusion
References
Using Differential Evolution to Estimate Labeler Quality for Crowdsourcing
1 Introduction
2 A Differential Evolution-Based Weighted Consensus Method in Crowdsourcing
3 Experiments and Results
4 Conclusions and Future Work
References
A Search Optimization Method for Rule Learning in Board Games
1 Introduction
2 Related Work
3 Preliminaries
3.1 General Game Playing and Game Description Language
3.2 Rule Learning
4 Learning Search Rules
5 Experiment Results
6 Conclusion and Future Work
References
Image Segmentation Based on MRF Combining with Deep Learning Shape Priors
Abstract
1 Introduction
2 Image Segmentation Based on MRF Combining with Deep Learning Shape Priors
2.1 MRF Image Segmentation
2.2 Appearance Priori
2.3 Deep Learning Shape Prior
3 Experimental Results and Analysis
4 Conclusion
References
An Automated Matrix Profile for Mining Consecutive Repeats in Time Series
1 Introduction
2 Related Work and Limitations
3 Problem Statement
4 Experimental Evaluation
5 Conclusion
References
High-Resolution Depth Refinement by Photometric and Multi-shading Constraints
1 Introduction
2 Our Approach
2.1 The Model
2.2 Optimization
2.3 The Algorithm and Implementation Details
3 Experiments
3.1 Setup
3.2 Quantitative Comparison
4 Conclusion
References
Weakly-Supervised Object Localization by Cutting Background with Deep Reinforcement Learning
1 Introduction
2 Methodology
2.1 Overview
2.2 Deep Reinforcement Learning for Localization
2.3 Refinement
3 Experiment
3.1 Network Training
3.2 Localization Prediction Metric
3.3 Performance and Analysis
4 Conclusion
References
Nature-Inspired Computational Model for Solving Bi-objective Traveling Salesman Problems
1 Introduction
2 Related Work
2.1 Basic Concepts of MOOP and Pareto-Optimal Solutions
2.2 The Definition of BTSP
3 Physarum-inspired NSGA_ii for BTSP
3.1 The Formulation of GA-based BTSP Method
3.2 The Formulation of pNSGA_ii
4 Experiments
4.1 Datasets and Parameters
4.2 Experimental Results
5 Conclusions
References
Differential Evolution-Based Weighted Majority Voting for Crowdsourcing
1 Introduction
2 Differential Evolution-Based Weighted Majority Voting
3 Experiments and Results
4 Conclusions
References
Scalable Machine Learning Techniques for Highly Imbalanced Credit Card Fraud Detection: A Comparative Study
Abstract
1 Introduction
2 Contribution
3 Review of Current Credit Card Fraud Detection
4 Experiments
4.1 Dealing with Imbalanced Data
4.2 Performance Metrics
4.3 Experimental Setup
4.4 Results and Discussions
5 Conclusion
References
View Decomposition and Adversarial for Semantic Segmentation
1 First Section
1.1 Introduction
2 Related Work
3 Approach
4 Experiments
4.1 Implementation Details and Baselines
4.2 Datasets
4.3 Fusion Strategy
4.4 Visual Analysis
5 Conclusion
References
Efficient Bayesian Optimisation Using Derivative Meta-model
1 Introduction
2 Related Background
2.1 Bayesian Optimisation
3 Framework
3.1 Meta-model
3.2 BO with Estimated Derivatives
3.3 Degree Estimation
4 Experiments
4.1 Experiment with Benchmark Test Functions
4.2 Hyperparamter Tuning
5 Conclusion
References
Prior Knowledge Guided Gene-Disease Associations Prediction: An Enhanced Inductive Matrix Completion Approach
Abstract
1 Introduction
2 Method
2.1 Inductive Matrix Completion
2.2 Enhanced Inductive Matrix Completion with Both Prior Sparse and Manifold Regularizations
2.3 Optimizing EIMC Using ADMM
3 Results
3.1 Dataset
3.2 Evaluation Method
3.3 Overall Performance
3.4 New Genes and New Diseases
4 Conclusion
Acknowledgements
References
Text Classification with Enriched Word Features
1 Introduction
2 Model Description
2.1 Enriched Word Representations
2.2 Word Feature Reuse
3 Datasets and Experimental Setup
3.1 Implementation Details
3.2 Baseline Models
4 Results and Discussion
4.1 Learned Word Representations
5 Conclusions
References
Attention-Based Linguistically Constraints Network for Aspect-Level Sentiment
1 Introduction
2 Attention-Based Linguistically Constraints LSTM Network (ALC-LN)
2.1 Linguistically Constraints
3 Experiment
3.1 Comparison with Baseline Methods
3.2 Analysis of ALC-LN Model
4 Related Work
5 Conclusion
References
Personalized POIs Travel Route Recommendation System Based on Tourism Big Data
Abstract
1 Introduction
2 Related Works
3 Research Methodologies
3.1 System Overview
3.2 Multi-source Tourism Big Data Integrating Module POIs Knowledgebase Creation
3.3 POI-Visit Sequences Mining Module
3.4 POIs Travel Route Recommendation Module
4 Experiment and Discussion
4.1 Data Set and Experiment Settings
4.2 Validation Experiment
4.3 Performance Experiment
5 Conclusions and Future Work
Acknowledgement
References
Analysing TV Audience Engagement via Twitter: Incremental Segment-Level Opinion Mining of Second Screen Tweets
1 Introduction
2 Incremental Segment Classification
2.1 Segment Classification Evaluation
3 Aggregated Audience Opinion
3.1 Polarizing and Volatile Panellists
3.2 Controversial Segments
3.3 Bias in Selection of Broadcast Tweets
4 Conclusion
References
Absolute Orientation and Localization Estimation from an Omnidirectional Image
1 Introduction
2 Approach
2.1 Spherical Panoramic Imaging Model
2.2 Estimating Orientation
2.3 Estimating Translation
3 Experiment
3.1 Dataset
3.2 Pose Accuracy
3.3 Time Cost
3.4 Visual Inspection
4 Conclusion
References
An Adaptive Clustering Algorithm by Finding Density Peaks
1 Introduction
2 Proposed Algorithm
2.1 Detecting Density Peaks Adaptively
2.2 Adjusting Strategy
2.3 Assignment Strategy
2.4 Merging Strategy
2.5 Main Steps of Proposed Algorithm
3 Experiments and Analyses
3.1 Descriptions to Datasets
3.2 Experimental Results and Analyses
4 Conclusions
References
Statutes Recommendation Using Classification and Co-occurrence Between Statutes
Abstract
1 Introduction
2 Related Work
3 Approach
3.1 Legal Background
3.2 Overview
3.3 Predicting CoAs
3.4 Recommending Statutes
3.5 Resorting Statutes
4 Experiment
4.1 Data Collection
4.2 Architecture Details
4.3 Experiments and Results
5 Conclusion
Acknowledgment
References
Robust and Real-Time Face Swapping Based on Face Segmentation and CANDIDE-3
1 Introduction
2 Related Work
2.1 Face Swapping
2.2 Face Modelling
2.3 Face Segmentation
3 System Design
3.1 Face Alignment and Modelling
3.2 Face Segmentation
3.3 Face Swapping and Blending
4 Experiments
4.1 Face Segmentation Results
4.2 Face-Swapping Results
5 Conclusion
References
Determining the Applicability of Advice for Efficient Multi-Agent Reinforcement Learning
1 Introduction
2 Joint Determination of Applicability of Advice
2.1 Generating Advised Action
2.2 Determining Whether the Advised Action Is Helpful
2.3 Determining Whether Advised Action Is Acceptable
3 Experimental Evaluation
3.1 Experimental Setup
3.2 Experimental Results and Analysis for Different Objectives
3.3 Experimental Results and Analysis in Dynamic Environment
4 Conclusion
References
Multi-object Detection Based on Deep Learning in Real Classrooms
1 Introduction
2 Related Work
2.1 Region Proposal Algorithms
2.2 RoI Pooling and RoIAlign Layers
3 Our Method
3.1 Feature Pyramid Network with RPNs
3.2 Position-Sensitive Feature Map and Position-Sensitive RoIAlign
3.3 Training
4 Experiments
4.1 Our Dataset
4.2 Experimental Details
4.3 Detection Results
5 Conclusion
References
Deep CRF-Graph Learning for Semantic Image Segmentation
1 Introduction
2 Learn CRF Graphs
2.1 Obtain Groundtruth from Pixel Labelling
2.2 Training
2.3 CRF-Graph Fusion
3 CRF Representation
4 Experimental Results
4.1 Results on KITTI
4.2 Results on PASCAL VOC 2012
5 Conclusion
References
Unrest News Amount Prediction with Context-Aware Attention LSTM
1 Introduction
2 GDELT Dataset
3 Notation and Problem Statement
4 Model
4.1 LSTM Encoder
4.2 Attention Layer
4.3 Context-Aware Layer
5 Experiments
5.1 Experimental Setting
5.2 Results
6 Related Work
7 Conclusion
References
Image Captioning with Relational Knowledge
1 Introduction
2 Related Work
3 Proposed Model
3.1 Overall Framework
3.2 Relational Knowledge with Word Representations
3.3 Join Relational Knowledge with Captioning Model
4 Experimental Results and Discussion
4.1 Datasets
4.2 Evaluation
5 Conclusion
References
An Elite Group Guided Artificial Bee Colony Algorithm with a Modified Neighborhood Search
Abstract
1 Introduction
2 Original ABC
3 Our Approach
3.1 Motivations
3.2 An Elite Group Guided Multi-dimension Search Strategy
3.3 A Modified Neighborhood Search Operator
3.4 Pseudo-code of ENABC
4 Experimental Verifications
4.1 Benchmark Functions and Parameter Settings
4.2 Comparison with Other ABC Variants
5 Conclusions
Acknowledgments
References
Exploiting Spatiotemporal Features to Infer Friendship in Location-Based Social Networks
1 Introduction
2 Related Work
3 The Proposed Framework
3.1 Preliminaries
3.2 Spatiotemporal Features to Infer Friendship(STIF)
4 Spatiotemporal Features
4.1 Exploiting Fine-Grained Temporal Features
4.2 Modeling Weekday and Weekend Check-ins
4.3 Measuring the Fine-Grained Location Weight
4.4 Co-occurrence Features
5 Experiments
5.1 Dataset
5.2 Performance Measures and Model Evaluation Method
5.3 Baseline Methods
5.4 Experiment Result Analysis
6 Conclusion
References
A Subsequent Speaker Selection Method for Online Discussions Based on the Multi-armed Bandit Algorithm
1 Introduction
2 Subsequent Speaker Determination Method Based on the Multi-armed Bandit Algorithm
2.1 Applying the Bandit Algorithm to Speaker Determination
2.2 Discussion Score
2.3 Clustering Using Bipartite Graph
3 Experiments
3.1 Experimental Settings
3.2 Results
4 Conclusion
References
An Entropy-Based Class Assignment Detection Approach for RDF Data
1 Introduction
2 Related Work
3 The Architecture of Class Assignment Detector
3.1 Module 1: Class Features Extraction
3.2 Module 2: Instance-Class Relationship Analysis
4 Experiments and Analysis
5 Conclusions and Future Work
References
Weighted Double Deep Multiagent Reinforcement Learning in Stochastic Cooperative Environments
1 Introduction
2 Preliminaries
3 Weighted Double Deep Q-Networks
4 Experiments
5 Conclusion
References
Automatically Classifying Chinese Judgment Documents Using Character-Level Convolutional Neural Networks
Abstract
1 Introduction
2 Related Work
3 Approach
3.1 Overview
3.2 Text Preprocessing
3.3 Classifier
4 Evaluation
4.1 Dataset
4.2 Experiments and Results
5 Conclusion and Future Work
Acknowledgment
References
RC-CNN: Reverse Connected Convolutional Neural Network for Accurate Player Detection
1 Introduction
2 Related Work
2.1 Player Detection
2.2 CNN-Based Object Detection
3 Methods
3.1 The Baseline Model: SSD
3.2 The RC-CNN Model
4 Experiments
4.1 Soccer Player Dataset
4.2 KITTI Pedestrian Dataset
5 Conclusion
References
Uncertainty Estimation for Strong-Noise Data
1 Introduction
2 Our Method
2.1 The Architecture
2.2 The Loss Function
2.3 Implement Details of Uncertainty Estimation
3 Experiments and Results
3.1 Fashion MNIST
3.2 Fashion MNIST with Gaussian Noise
3.3 Daily Data of NYSE Stocks
4 Conclusion and Future Work
References
Reciprocal Ranking: A Hybrid Ranking Algorithm for Reciprocal Recommendation
Abstract
1 Introduction
2 Problem Definition
3 Reciprocal Ranking
3.1 Assumptions
3.2 Objective Function
3.3 Model Learning
4 Experiment
4.1 Data
4.2 Evaluation Metrics
4.3 Result and Analysis
5 Conclusion
Acknowledgments
References
Robust Low-Rank Recovery with a Distance-Measure Structure for Face Recognition
1 Introduction
2 Robust Low-Rank Representation Algorithm (RLRR)
3 The Optimization of RLRR
4 Classification Method
5 Experiments
5.1 Experiments on AR Database
5.2 Experiments on LFW Database
5.3 Experiments on Extended Yale B Database
6 Conclusions
References
A Surface Defect Detection Method Based on Positive Samples
Abstract
1 Introduction
2 Related Work
2.1 Defect Repair Model Based on Positive Samples
2.2 Autoencoder
3 Method
3.1 Objective
3.2 Network Structure and Artificial Defects
3.3 To Get the Position of the Defect
4 Experiment
4.1 Preparation
4.2 Result
5 Conclusion
Acknowledgments
References
Joint Multi-field Siamese Recurrent Neural Network for Entity Resolution
1 Introduction
2 Related Work
3 Our Method
3.1 Problem Definition
3.2 Joint Multi-field Siamese Recurrent Neural Network
4 Experiments
4.1 Datasets
4.2 Comparison with Baseline Methods
4.3 Effectiveness of Different Network Components
5 Conclusion
References
Using Machine Learning for Determining Network Robustness of Multi-Agent Systems Under Attacks
1 Introduction
2 Preliminary
3 The Method
3.1 Preprocessing Steps
3.2 Spectral Clustering Steps
3.3 Network Robustness Classifier Details
4 Experiment
4.1 Datasets and Compared Methods
4.2 Experiment Setting
4.3 Experiment Results
5 Conclusion
References
Collective Hyper-heuristics for Self-assembling Robot Behaviours
1 Introduction
2 Methodology
3 Self-assembling Swarm Robot Cleaner
3.1 Implementation
3.2 Heuristic Repository
4 Experiments and Results
4.1 Robustness in Different Environments
4.2 Comparison of Collective Decision Making Stategies
5 Conclusions and Future Works
References
Matrix Factorization for Identifying Noisy Labels of Multi-label Instances
1 Introduction
2 Related Work
3 Proposed Method
3.1 Feature-Based Regularization
3.2 Label-Based Regularization
4 Experimental Setup
5 Experimental Results and Analysis
5.1 Noisy Label Identification
5.2 Parameter Sensitivity Analysis
6 Conclusions and Future Work
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken