New Frontiers in Artificial Intelligence: JSAI-isAI 2021 Workshops, JURISIN, LENLS18, SCIDOCA, Kansei-AI, AI-BIZ, Yokohama, Japan, November 13–15, ... (Lecture Notes in Artificial Intelligence)

Preface
Organization
Contents
JURISIN 2021
	Fifteenth International Workshop on Juris-informatics (JURISIN 2021)
		1 The Workshop
		2 Papers
Prediction Model for Drunk Driving Sentencing: Applying TextCNN to Chinese Judgement Texts
	1 Introduction
	2 Literature Review and Research Design
	3 Data Source and Processing
		3.1 Data Source
		3.2 Input Data Processing
		3.3 Choosing Classification Method
	4 Model Training and the Result
		4.1 Model Training
		4.2 Presentation of Model Results
	5 Discussion and Conclusion
	References
User-Guided Machine Understanding of Legal Documents
	1 Introduction
	2 Overview of the Smart Document Editor (SDE)
	3 Ontology Discovery Process
	4 Modelling and Testing the Logic Model
	5 Formal Correspondence
	6 ASP Representations
	7 Evidence of Usability
	8 Evaluation and Future Work
	9 Conclusion
	References
Benchmarks for Indian Legal NLP: A Survey
	1 Need for Indian Legal NLP Benchmarks
	2 How Can NLP Benchmarks Help Innovation in Indian Legal System?
	3 Data Availability
		3.1 Court Judgments
		3.2 First Investigation Reports
		3.3 Indian Laws
	4 Indian Legal NLP Benchmark Areas
		4.1 Information Extraction
		4.2 Summary Generation
		4.3 Case Similarity
		4.4 Reasoning Based NLP
	5 Operationalizing Indian Legal NLP Benchmarks
	References
Computer-Aided Comparative Law on Meiji Civil Code
	1 Introduction
	2 Problem Definition and Modeling
		2.1 Reference Relation Estimation Problem
		2.2 Algorithm for Reference Relation Estimation
		2.3 Finding the Best Distance Measure
		2.4 Definition of Distance
	3 Evaluation Experiment
		3.1 Target Codes
		3.2 Preprocessing
		3.3 Result
	4 Discussion
		4.1 Macro Perspective
		4.2 Micro Perspective
	5 Conclusions
	References
Topic Modelling for Risk Identification in Data Protection Act Judgements
	1 Introduction
	2 Background
	3 Data
	4 Methods
		4.1 Structural Topic Modelling
		4.2 Determining Topic Numbers
		4.3 Pre-processing of Text
	5 Results
		5.1 Data Protection Case Topics
		5.2 Court Outcomes by Topic
	6 Discussion
		6.1 Associating Outcomes with Contexts
		6.2 The Utility of STM for Legal Analysis
		6.3 Limitations
	7 Conclusions
	References
A Compliance Mechanism for Planning in Privacy Domain Using Policies
	1 Introduction
	2 Backgrounds
		2.1 The SPECIAL Policy Language
		2.2 Event Calculus
	3 Modular Framework
		3.1 A Use Case Model
		3.2 Planning
		3.3 Compliance Engine
	4 Evaluation
	5 Related Works
	6 Conclusion and Future Work
	References
LENLS 18
	Logic and Engineering of Natural Language Semantics (LENLS) 18
Remembering Individuals and Remembering Scenes
	1 Introduction
	2 Not a `Classical' Ambiguity
		2.1 Not a Scope Ambiguity
		2.2 Not a Structural Ambiguity
	3 Other Properties of `rememberDP'-Reports
		3.1 `Individual Only'-Interpretations
		3.2 Cross-Categorial Entailments
	4 Previous Accounts
	5 Proposal
		5.1 Capturing the Individual/Scene-Ambiguity
		5.2 Predicting `Individual Only'-Interpretations
		5.3 Capturing Referential Opacity
	6 Outlook
	References
Factivity Variation in Episodic Memory Reports
	1 Introduction
	2 Existing Accounts
	3 Background: Experiential Parasitism
	4 A Uniform Semantics for Episodic Remember
		4.1 Capturing Experientiality Inferences
		4.2 Capturing Factivity Inferences
		4.3 Reports of False Episodic Memories
	5 Outlook
	References
The Duality of Negative Attitudes in Japanese Conditionals
	1 Introduction
	2 Negative Attitudes of Mono-nara Conditionals
	3 Proposal and Analysis
		3.1 The Meaning of Mono-nara
		3.2 [Y]oo as a Weak Necessity Modal with a Flexible Flavor
		3.3 Composing the Negative Attitudes
	4 Predictions and Implications
		4.1 Interaction with Possibility Modals
		4.2 Compatibility with Counterfactuals
		4.3 Negation and Licensing of Minimizers
	5 Conclusion
	References
Interpretations of Sense-Based Minimizers in Japanese and English: Direct and Indirect Sense-Based Measurements
	1 Introduction
	2 The Meaning of Japanese kasukani `faintly'
		2.1 The Experiential Component of kasukani
		2.2 The Barely-Component of kasukani
		2.3 The Non-at-issue (CI) Property of kasukani
		2.4 Formal Analysis of kasukani
	3 Indirect Measurement: Measuring the Degree of Emotion via Perception
	4 English faintly
	5 Conclusion
	References
Presuppositions and Comparison Classes in Japanese hodo-equatives
	1 Introduction
	2 Quandary
	3 The Semantics of hodo and mo
		3.1 The Semantics of hodo
		3.2 The Semantics of mo ``even/also''
	4 Proposal
	5 Conclusion
	References
The Semantics and Pragmatics of the Japanese Honorific Titles San, Kun, and Chan
	1 Introduction
	2 Affixal Designation Terms
	3 San, Kun, and Chan: Basic Facts
	4 Formal Semantics of Honorification
	5 The Semantics and Pragmatics of San
	6 The Semantics and Pragmatics of kun
		6.1 Kun applied to males
		6.2 Gender-neutral kun
	7 The Semantics and Pragmatics of chan
		7.1 Chan applied to children and females
		7.2 Gender/age-neutral chan
	8 Conclusion
	References
Honorifics, Grounds and Ideologies
	1 Introduction
	2 Honorifics and Their Semantics
	3 Ideologies
	4 Honorific Grounds via Ideologies
	5 Grounds as Category
	6 Other Examples
	7 Conclusion and Prospects
	References
QNP Textual Entailment with Polynomial Event Semantics
	1 Introduction
	2 Generalized Quantifiers
		2.1 Negative Quantification and Downward Monotonicity
		2.2 Many, Most, Few
	3 Algebra of Polynomial Event Semantics
		3.1 Negation
	4 Copula Clauses
	5 Existence and Subject Relative Clauses
	6 Related Work
	7 Conclusions
	References
Parsed Corpus Development with a Quick Access Interface
	1 Introduction
	2 The Corpus Overview Page and Interface Map
	3 Accessing Content by Providing ID Information
	4 The Underlying Data Format
	5 Search Pages
		5.1 Grep Page
		5.2 TGrep2 Page
	6 Visualisation Pages
		6.1 Tree Page
		6.2 Word Page
		6.3 Graph Page
	7 A Final Example
	8 Closing Remarks
	References
A Proof-Theoretic Analysis of Weak Crossover
	1 Quantificational Weak Crossover
		1.1 Background
		1.2 The Right Prediction
		1.3 The Wrong Prediction
	2 Proposal: Underspecified Types in DTS
	3 Global and Local Accommodations
	4 Conclusion
	A Dependent Type Theory
		A.1 Syntax
		A.2 Type System
		A.3 Type Checking
		A.4 Type Inference
	References
Probabilistic Compositional Semantics, Purely
	1 Introduction
	2 Formal Semantics
	3 The Traditional Interpretation
	4 The Probabilistic Interpretation
	5 Probabilistic Programs
	6 Bayesian Inference
		6.1 Semantic Learning
		6.2 RSA: Background
		6.3 RSA: Implementation
	7 Conclusion
	References
Pluralism for Relativists: A New Framework for Context Dependence
	1 Introduction
	2 Context-Sensitivity
	3 Relativisms
		3.1 Non-indexical Contextualism
		3.2 Assessment-Sensitivity
	4 Retraction
		4.1 Non-indexical Contextualism vs Assessment-Sensitivity
		4.2 Variance to Remedy Strength
	5 Pluralism
		5.1 Motivations
		5.2 Pluralism for Variance
		5.3 Formalizing Initialization
		5.4 The Model
	6 Taking Stock
	References
Cheap Talk Under Partial Conflicts: A Dynamic Analysis of Pragmatic Meaning
	1 Introduction
	2 Communication Under Partial Conflicts: Inference and Credibility
	3 CS Game with the Expanded Signal Structure
	4 Iterated Best Response (IBR) Under the Expanded Signal Structure
	5 Conclusion
	References
SCIDOCA 2021
	Fifth International Workshop on SCIentific DOCument Analysis (SCIDOCA2021)
	1 The Workshop
	2 The Papers
Investigating the Effects of Pre-trained BERT to Improve Sparse Data Recommender Systems
	1 Introduction
	2 Our Model
		2.1 Probabilistic Matrix Factorization (PMF)
		2.2 Representations of Item Reviews
	3 Experiments
		3.1 Data
		3.2 Baseline
		3.3 Metrics
		3.4 Settings
		3.5 Results
		3.6 Analyses and Discussions
	4 Conclusion
	References
A Novel Pipeline to Enhance Question-Answering Model by Identifying Relevant Information
	1 Introduction
	2 Related Works
		2.1 Machine Reading Comprehension
		2.2 Information Retrieval
		2.3 Question-Answering Context Reduction
	3 The Proposed System
		3.1 Overall
		3.2 Potential Sentence Classification Model
		3.3 Answer Extraction
	4 Experiments and Results
		4.1 Dataset
		4.2 Models
		4.3 Experimental Results
	5 Discussion
	6 Conclusion
	References
KANSEIAI 2021
	Kansei and Artificial Intelligence 2021
	1 The Workshop
	2 Paper
The Influence of Visual Context on the Naturalness Impression of Auditory Stimuli
	1 Introduction
	2 Evaluation Experiment 1: Influence of Images on the Monitor on Sound Evaluation
		2.1 Subjects
		2.2 Experimental Conditions
		2.3 Experimental Procedure
	3 Evaluation Experiment 2: Influence of Indoor Environment on Sound Evaluation
		3.1 Subjects
		3.2 Experimental Conditions
		3.3 Experimental Procedure
	4 Experimental Results
		4.1 Analysis of Scale and Response Results by SD Method
		4.2 Analysis of the Results of Quantifying Onomatopoeia with the System
		4.3 Analysis of the Number of Phonemes of the Onomatopoeia that Appeared
	5 Discussion
		5.1 Differences in Ratings Between Adjective-Scale and Onomatopoeic Responses
		5.2 Phonology of the Onomatopoeia Responses
	6 Conclusion
	References
AI-Biz 2021
	Artificial Intelligence of and for Business (AI-Biz2021)
		1 The Workshop
		2 Papers
		3 Acknowledgment
Legality Identification of Japanese Online Advertisements Using Complex-Valued Support Vector Machines with DFT-Coded Document Vectors
	1 Introduction
	2 Relatated Work
	3 Legality of Advertising Documents
		3.1 Definition of Problematic Documents
	4 Features of Inappropriate Advertising Documents
		4.1 Frequency Features of Words
		4.2 Part-of-Speech Features
		4.3 Documents that Cannot Be Discriminated by Words Only
	5 Features to Discriminate Advertising Documents
		5.1 Properties of Effective Features for Discrimination
		5.2 Document Vector Combining Uw and DFT
	6 Discriminating Documents Using Complex-Valued Support Vector Machine
	7 Discrimination Simulation of Cosmetic Advertisements
		7.1 Performance Indicators of Discriminant Model
		7.2 Simulation Using Holdout Method
		7.3 Discriminant Models and Document Vectors to Compare
		7.4 Simulation Results
	8 Discussion
	9 Conclusion
	References
What is the Investment Strategy to Overcome the Severe Business Environment?
	1 Introduction
	2 Setting the Condition
	3 Modeling
		3.1 Composition of the Decision-Making Model in the Agent
		3.2 Agent Details
		3.3 Defining the Relationship Between the Environment and the Agent
		3.4 Environment Details
		3.5 Simulation Overview
	4 Verification Results
		4.1 Hiring Rates and Firing Rates
		4.2 Change in Investment Rate and Skills of Initial Number of People
		4.3 Discussion
	5 Conclusions and Further Work
	References
Empirical Analysis of the Impact of Mergers and Acquisitions on Innovation Activities Through Clustering
	1 Introduction
	2 Sample and Data
	3 Methodology
		3.1 Measurement of Innovation Output
		3.2 Classification Method Through Clustering
		3.3 Measurement of Stock Price Return
		3.4 Calculation Method of Patent Document Vector
		3.5 Measurement of Technological Distance
	4 Empirical Result
		4.1 Classification Results Through Clustering
		4.2 Scale of Innovation Output
		4.3 Average Stock Price Return
		4.4 Technological Distance
		4.5 Verification of Analysis Results
	5 Discussion
	6 Conclusion
	References
Assessment of the Impact of COVID-19 Infections Considering Risk of Infected People Inflow to the Region
	1 Introduction
	2 Related Work
	3 SEIR Infection Model Considering Inflow Risk
	4 Sapporo Model Construction and Experimental Results
	5 Construction and Experimental Results of the Individual-Based Model of Tokyo Suburbs
	6 Discussion
	7 Summary
	References
Author Index