
Advances in Artificial Intelligence
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
This book presents selected and extended papers from the largest conference on artificial intelligence in Japan, which was expanded into an internationalized event for the first time in 2019: the 33rd Annual Conference of the Japanese Society for Artificial Intelligence (JSAI 2019), held on June 4-June 7, 2019 at TOKI MESSE in Niigata, Japan.
The book's content has been divided into six major sections, on (I) knowledge engineering, (II) agents, (III) education and culture, (IV) natural language processing, (V) machine learning and data mining, and (VI) cyber physics.
Given its scope, the book offers a valuable reference guide for professionals, undergraduate and graduate students engaged in disciplines, fields, technologies, or philosophies relevant to AI, e.g., computer/data science, robotics, linguistics, and physics, introducing them to recent advances in this area and discussing the human society of tomorrow.
More details
Other editions
Additional editions

Content
- Intro
- Preface
- Contents
- Knowledge Engineering
- Using Sequence Constraints for Modelling Network Interactions
- 1 Introduction
- 2 Behavioural Constraints for Sequence Interactions
- 2.1 Constraint Set
- 2.2 Mining the Patterns
- 2.3 Applications
- 3 Application
- 3.1 Data
- 3.2 Interpretation
- 4 Conclusion and Future Work
- References
- Prediction of Onset of Lifestyle-Related Diseases Using Regular Health Checkup Data
- 1 Introduction
- 2 Target Data
- 2.1 Data Overview
- 2.2 Disease Names as Prediction Targets
- 2.3 Feature Values for Prediction
- 2.4 Characteristics of Data
- 3 Data Selection and Machine Learning
- 3.1 Data Selection
- 3.2 Machine Learning for Imbalanced Data
- 4 Results and Discussion
- 4.1 Comparison with a Baseline Method
- 4.2 Important Features and Influence of Disease Frequency
- 4.3 Comparison with an Oversampling Method
- 5 Conclusion and Future Work
- References
- Feasible Affect Recognition in Advertising Based on Physiological Responses from Wearable Sensors
- 1 Introduction
- 2 Data Acquisition and Preprocessing
- 2.1 Electroencephalogram (EEG)
- 2.2 Electrocardiogram (ECG)
- 2.3 Eye-Tracking
- 3 Research Methodology
- 3.1 Data Trends
- 3.2 Window Recognition
- 3.3 Sequence Learning
- 4 Experiments and Results
- 4.1 Preliminary Experiment
- 4.2 Recognition Experiment
- 5 Discussion and Conclusion
- References
- Deep Markov Models for Data Assimilation in Chaotic Dynamical Systems
- 1 Introduction
- 2 Deep Markov Model
- 2.1 Generative Model
- 2.2 Inference Network
- 3 Experiments and Results
- 3.1 Dataset and Evaluation
- 3.2 Experiment Modifications
- 3.3 Results
- 4 Conclusion
- References
- Reducing the Number of Multiplications in Convolutional Recurrent Neural Networks (ConvRNNs)
- 1 Introduction
- 2 ConvRNNs
- 2.1 Basic ConvRNN Architecture
- 2.2 Next Video Frame Prediction Model Using ConvRNN
- 3 Proposed Improvements
- 4 Experimental Results
- 4.1 Database Description
- 4.2 Performance Evaluation
- 5 Conclusions
- References
- Agents
- Rewards Visualization System Promotes Information Provision
- 1 Introduction
- 2 Models and Methods
- 2.1 A Restricted Meta Reward Game Model
- 2.2 Adaptive Process of Strategy
- 2.3 Simulation Scenarios
- 2.4 Parameter Setting
- 3 Simulation Results
- 3.1 Summary of Comparing Three Scenarios
- 3.2 Effects on Cooperation of the Rate of Browseable Users
- 3.3 Adaptation of Reference Strategy
- 4 Conclusion
- References
- An Automated Negotiating Agent that Searches the Bids Around Nash Bargaining Solution to Obtain High Joint Utilities
- 1 Introduction
- 2 Negotiation Environment
- 2.1 Negotiation Competition
- 2.2 Preferences of the Negotiating Agents
- 3 Proposed Approach
- 3.1 Bid Searching Strategy
- 3.2 Bid Acceptance Strategy
- 4 Results and Evaluation
- 5 Related Works
- 6 Conclusion and Future Work
- References
- ANAC 2018: Repeated Multilateral Negotiation League
- 1 Introduction
- 2 ANAC 2018 Competition Challenges
- 3 Competition Setup
- 4 Result of the Competition
- 4.1 Qualification Results
- 4.2 Final Results: Individual Utility Category
- 4.3 Social Welfare Category
- 4.4 Further Analysis of ANAC 2018 Winners
- 5 Related Work
- 6 Conclusion
- References
- Flexibility of Emulation Learning from Pioneers in Nonstationary Environments
- 1 Introduction
- 2 Reinforcement Learning Algorithms
- 2.1 Q-Learning
- 2.2 Inverse Reinforcement Learning as Imitation
- 2.3 Emulation by Risk-Sensitive Satisficing (RS)
- 3 First Experiment
- 3.1 Task: SwitchWorld
- 3.2 Task: UnsteadySwitchWorld
- 3.3 Simulation and Result
- 3.4 Discussion
- 4 Second Experiment
- 4.1 Task: LoopSwitchWorld and LoopUnsteadySwitchWorld
- 4.2 Simulation and Result
- 4.3 Discussion
- 5 General Discussion
- 5.1 RS+GRC in Terms of Spatial and Temporal Efficienty
- 5.2 Adaptation to Non-stationary Environments
- 6 Conclusion
- References
- Education and Culture
- Computerized Adaptive Testing Method Using Integer Programming to Minimize Item Exposure
- 1 Introduction
- 2 Item Response Theory
- 3 Computerized Adaptive Testing
- 3.1 Constrained CAT with Item Exposure Control
- 4 Proposed Method
- 5 Numerical Evaluation
- 5.1 Simulation Experiment
- 5.2 Experiment Conducted Using Actual Data
- 6 Conclusions
- References
- Proposal of Context-Aware Music Recommender System Using Negative Sampling
- 1 Introduction
- 2 Related Work
- 2.1 Implicit Feedback
- 2.2 Context Information
- 3 Proposed Method
- 3.1 Context-Aware Recommendation
- 3.2 User-Oriented Negative Sampling
- 4 Experiment
- 4.1 Dataset
- 4.2 Experimental Setup
- 4.3 Evaluation Metric
- 4.4 Result and Discussion
- 5 Conclusion and Future Work
- References
- The Influence of Story Creating Activities While Appreciating Abstract Artworks
- 1 Introduction
- 1.1 Abstract and Representational Paintings
- 1.2 Creating Story About Abstract Paintings
- 1.3 Explaining Abstract Paintings
- 2 Materials and Methods
- 2.1 Participants
- 2.2 Stimuli
- 2.3 Conditions
- 2.4 Hypothesis
- 3 Results and Discussion
- 3.1 Similar Stories for an Artwork?
- 3.2 Settings: Place
- 3.3 Settings: Time
- 3.4 Discussions
- 4 Conclusions
- References
- Natural Language Processing
- Unsupervised Joint Learning for Headline Generation and Discourse Structure of Reviews
- 1 Introduction
- 2 Related Research
- 2.1 Supervised Review Summary Generation
- 2.2 Un-/Semi-supervised Summarization
- 2.3 Discourse Parsing and Its Application
- 3 Proposed Model
- 3.1 Model Training
- 3.2 Headline Generation
- 3.3 Marginal Probability of Dependency
- 4 Experiments
- 4.1 Dataset
- 4.2 Experimental Details
- 4.3 Baseline
- 4.4 Results
- 5 Discussion
- 6 Conclusion
- References
- Extraction of Online Discussion Structures for Automated Facilitation Agent
- 1 Introduction
- 2 Related Works
- 2.1 Argumentation Mining
- 2.2 Japanese Natural Language Processing
- 3 Extracting IBIS Structures with Deep Learning
- 3.1 Node Extraction
- 3.2 Link Extraction
- 4 Experiments
- 4.1 Dataset
- 4.2 Experimental Settings
- 5 Results and Discussion
- 5.1 Results
- 5.2 Discussion
- 6 Conclusion and Future Work
- References
- Variables Extraction in Natural (English) Language Through Possessive Relationships
- 1 Introduction
- 2 Related Works
- 2.1 Variable Data and Predicate Logic
- 2.2 Entity Recognition
- 2.3 Data Jacket
- 3 Experiment
- 3.1 Procedures
- 4 Discussion and Future Work
- References
- GTransE: Generalizing Translation-Based Model on Uncertain Knowledge Graph Embedding
- 1 Introduction
- 2 Problem Definition
- 3 Related Work
- 4 Generalizing Translation-Based Model
- 5 Experiments and Results
- 5.1 Dataset
- 5.2 Experimental Setup
- 5.3 Results
- 6 Conclusion
- References
- Latent-Space Data Augmentation for Visually-Grounded Language Understanding
- 1 Introduction
- 2 Related Work
- 3 Problem Statement
- 4 Multimodal Target-Source Classifier Model with Data Augmentation
- 5 Experiments
- 6 Conclusion
- References
- Machine Learning and Data Mining
- A Community Sensing Approach for User Identity Linkage
- 1 Introduction
- 2 Related Work
- 2.1 Structure-Based Approaches
- 2.2 Embedding-Based Approaches
- 2.3 Multi-task Learning
- 3 Problem Definition
- 4 Community Sensing User Identity Linkage
- 4.1 Network Embedding
- 4.2 Community Clustering
- 4.3 Latent Space Mapping
- 5 Experiment
- 5.1 Data Preparation
- 5.2 Evaluation Metrics
- 5.3 Comparative Methods
- 5.4 Results
- 6 Conclusion and Future Work
- References
- Social Influence Prediction by a Community-Based Convolutional Neural Network
- 1 Introduction
- 2 Related Work
- 2.1 Diffusion Models
- 2.2 Prediction of Information Spread Size
- 3 Methodology
- 3.1 Social Influence Learning on Deep Neural Network
- 3.2 Social Influence Learning on Community-Based Convolutional Neural Network
- 4 Experiments
- 4.1 Dataset
- 4.2 Compared Methods
- 4.3 Evaluated Metrics
- 4.4 Results and Discussions
- 5 Conclusions and Future Works
- References
- Segment Information Extraction from Financial Annual Reports Using Neural Network
- 1 Introduction
- 2 Task Setting
- 3 Extraction of the Segment Information Using RNN
- 3.1 Structure of RNN
- 3.2 Segment Explanation Start Layer
- 3.3 Learning
- 3.4 Segment Information Extraction
- 4 Experimental Evaluation
- 4.1 Dataset
- 4.2 Evaluation Metrics
- 4.3 Comparison Method
- 4.4 Other Settings
- 5 Results and Discussion
- 5.1 Result and Discussion
- 5.2 Task Difficulty in Extraction of the Segment Explanation
- 6 Related Works
- 7 Conclusion
- References
- One-Shot Learning Using Triplet Network with kNN Classifier
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Conditional Variational Autoencoders
- 3.2 Triplet Network
- 3.3 KNN Classifier
- 4 Experiment
- 4.1 Dataset
- 4.2 Triplet Selection
- 4.3 Results
- 5 Discussion
- References
- Trees Detection on Google Street View Images Using Deep Learning and City Open Data
- 1 Motivation and Research Background
- 2 Related Works
- 2.1 Trees Detection
- 2.2 Machine Learning Technologies
- 3 System Implementation
- 3.1 Using City Open Data to Find Isolated Trees' Positions, Then Collecting Their Pictures from Google Street View
- 3.2 Adopting Deep Learning Approaches for Trees Image Segmentation
- 3.3 Selecting Appropriate Deep Learning Models for Tree Segmentation
- 3.4 Data Augmentation
- 4 Current Experiment Results
- 5 Conclusion
- References
- Cyber Physics
- Classification of Accessibility and Synergy in the Tokyo Area Multimodal Transportation Network
- 1 Introduction
- 2 Socioececonomic Data
- 3 Network Construction
- 4 Methods
- 4.1 Measures of Accessibility
- 4.2 Measures of Synergy
- 4.3 K-Means Clustering
- 4.4 Feature Importance
- 5 Results 1: Accessibility Clusters
- 5.1 Public Transportation Accessibility
- 5.2 Integrated Transportation Accessibility
- 5.3 Accessibility Determining Factors
- 6 Results 2: Network Synergy Clusters
- 6.1 Public Transportation Synergy
- 6.2 Integrated Transportation Synergy
- 6.3 Synergy Determining Factors
- 7 Conclusions
- References
- Modelling Naturalistic Work Stress Using Spectral HRV Representations and Deep Learning
- 1 Introduction
- 1.1 HRV Stress Analysis
- 1.2 Deep Learning and Heart Rate Analysis
- 2 Methodology
- 2.1 Dataset
- 2.2 Signal Processing
- 3 Experiments and Validation
- 3.1 Dataset and Models
- 3.2 Results and Discussion
- 3.3 ROC Analysis
- 4 Summary and Future Work
- References
- Privacy-Preserving Resident Monitoring System with Ultra Low-Resolution Imaging and the Examination of Its Ease of Installation
- 1 Introduction
- 2 Indoor Monitoring System Using an Ultra-low-resolution Infrared Array Sensor
- 2.1 Posture Classification and Experimentation System
- 2.2 Posture Classification Using DCNN
- 2.3 Effects of Data Conversion on Classification Accuracy
- 3 Development of Learning Data-Generating Applications and Evaluation of Posture Classifications
- 3.1 Production of Pseudo-data Simulator for Data Learning
- 3.2 Evaluation of Data Learning Using Pseudo-Data
- 4 Effects of Data Conversions on Classification Accuracy
- 4.1 Evaluations on Installation Height
- 4.2 Evaluations of Sensor Angle
- 5 Discussion
- 6 Conclusion
- References
- Identity Verification Using Face Recognition Improved by Managing Check-in Behavior of Event Attendees
- 1 Introduction
- 2 Personal Authentication
- 3 Ticket ID System Using Face Recognition
- 3.1 Outline of Ticket ID System
- 3.2 Problems with One-Stop System
- 4 Continuous-Face-Recognition System
- 4.1 Managing Check-in Behavior of Attendees
- 4.2 Configuration of Proposed System
- 4.3 Identity-Verification Procedure
- 4.4 Operational Steps
- 5 Demonstration of Proposed System
- 5.1 Results
- 5.2 Discussion
- 6 Future Issues
- 7 Conclusion
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.