Information Integration and Web Intelligence: 25th International Conference, iiWAS 2023, Denpasar, Bali, Indonesia, December 4–6, 2023, Proceedings (Lecture Notes in Computer Science)

This book constitutes the refereed conference proceedings of the 25th International Conference on Information Integration and Web Intelligence, iiWAS 2023, organized in conjunction with the 21st International Conference on Advances in Mobile Computing and Multimedia Intelligence, MoMM2023, held in Denpasar, Bali, Indonesia, during December 4-6, 2023. The 24 full papers and 24 short papers presented in this book were carefully reviewed and selected from 96 submissions. The papers are divided into the following topical sections: business data and applications; data management; deep and machine Learning; generative AI; image data and knowledge graph; recommendation systems; similarity measure and metric; and topic and text matching. Preface Organization ABC (AI-Big Data Convergence) Forum Demystifying Large Language Models and Generative Pretrained Transformer (Tutorial) The Future of LLMs (Panel) Contents Business Data and Applications How Domain Engineering Can Help to Raise Adoption Rates of Artificial Intelligence in Healthcare*-12pt 1 Introduction 2 Domain Engineering 3 Applying Domain Engineering to Decision Support System Development 4 Conclusion References Enhancing AI Adoption in Healthcare: A Data Strategy for Improved Heart Disease Prediction Accuracy Through Deep Learning Techniques 1 Introduction 2 Related Work 3 Methodology 4 Results 5 Conclusion References WISHFUL - Website Extraction of Institutional Sources with Heterogeneous Factors and User-Driven Linkage 1 Introduction 2 Concept and Implementation 2.1 Locating the Main Content 2.2 Sorting the Content Based on Certain Patterns 2.3 Extraction of Raw Content 3 Evaluation 3.1 Effectiveness of the Proposed Web Extraction Approach 3.2 Usefulness of Chosen Source for Data Collection 4 Summary and Conclusion References DISA - A Blockchain-Based Distributed Information Security Audit 1 Introduction 2 Background and Related Work 2.1 Blockchain Technology 2.2 Related Work 3 Approach 3.1 Concept and Purpose 3.2 Data Structures and Operations 4 Experimental Prototype 4.1 Substrate Prototype 4.2 Results 5 Discussion 6 Conclusion References Analysing Online Review by Bank Employees: A Predictive Analytics Approach 1 Introduction 2 Related Work 3 Dataset and Empirical Results 3.1 Dataset 3.2 Modelling and Performance Evaluation 4 Discussion 5 Concluding Remarks References Optimizing Visit Booking at CERN: A Drools-Based Approach 1 Introduction 2 State of the Art 3 Implementation 3.1 Drools Rules 3.2 Rules Engine Input and Output 4 Results 5 Conclusion References A Label Aggregation Method Using Worker Quality in Crowdsourcing 1 Introduction 2 Related Work 3 Proposed Method 3.1 Model Construction 3.2 Calculation of Agreement Rates 3.3 Calculation the Weights of Workers' Votes 4 Experiments 4.1 Experimental Setup 4.2 Results and Discussion 5 Conclusion References ImputAnom: Anomaly Detection Framework Using Imputation Methods for Univariate Time Series 1 Introduction 2 Methodology 3 Experiments and Discussion 4 Conclusion and Future Work References Smart Cities as Hubs: Navigating the Smart City Data Providers Landscape 1 Introduction 2 The SCHub Concept 2.1 SCHub Architecture 2.2 Functional Requirements 2.3 Non-functional Requirements 3 Smart City Data Providers 4 Discussion and Future Research Directions References Revolutionizing Real Estate: A Blockchain, NFT, and IPFS Multi-platform Approach 1 Introduction 2 Related Work 3 Approach 3.1 Traditional Real Estate Management: An Overview 3.2 Proposed Architecture for Blockchain-Based Real Estate Transactions 4 Evaluation Scenarios 4.1 Transaction Fee Analysis 4.2 Gas Limit Analysis 5 Conclusion References Data Management Towards a Unified Symbolic AI Framework for Mining High Utility Itemsets 1 Introduction 2 General Setting 2.1 High Utility Intemset Mining Problem 2.2 The Satisfiability Problem in Propositional Logic 3 A Unified SAT-Based Framework for HUIM 4 Efficient Solving Strategy for HUIM Problem 5 Empirical Evaluation 6 Conclusion References RDF Data Partitioning for Efficient SPARQL Query Processing with Spark SQL 1 Introduction 2 Preliminaries 2.1 RDF and SPARQL 2.2 Processing RDF Using Spark SQL 3 Related Work 3.1 Workload-Based Methods 3.2 Methods Using Betweenness Centrality and Indexing 4 Proposed Method 4.1 Data Partitioning Process 4.2 Query Processing 5 Performance Study 5.1 Setup 5.2 Datasets 5.3 Performance Measures 5.4 Results 6 Conclusions References A Comparative Evaluation of Additive Separability Tests for Physics-Informed Machine Learning 1 Introduction 2 Background and Related Work 2.1 Additive Separability 2.2 Multilayer Perceptron Neural Network Surrogates 3 Methodology 4 Performance Evaluation 4.1 Experimental Setup 4.2 Experimental Results 5 Conclusion References Data Integration in a Multi-model Environment 1 Introduction 2 Functional Data Modelling 2.1 Functional Data Types 2.2 Typed Lambda Calculus 2.3 Conceptual Modelling with Attributes 2.4 Querying with Attributes 3 Multi-model Approach to Data Integration 4 Querying Multi-model Data 5 Conclusions References Character Entity Recognition Using Hybrid Binary-Particle Swarm Optimization and Conditional Random Field on Balinese Folklore Text 1 Introduction 2 Character Named Entity on Balinese Folklore Dataset 3 Hybrid Binary Particle Swarm Optimization and Conditional Random Field for Character Named Entity Recognition 4 Result and Discussion 5 Conclusion References On Observing Patterns of Correlations During Drill-Down 1 Introduction 2 Problem Statement 3 Experimental Study 3.1 CMDC Dataset 3.2 Preprocessing 3.3 Setup 4 Evaluation 5 Conclusion References Deep and Machine Learning Unsupervised Clustering and Explainable AI for Unveiling Behavioral Variations Across Time in Home-Appliance Generated Data 1 Introduction 2 Theoretical Background and Related Work 2.1 Harnessing Temporal Insights: Machine Learning Algorithms for Time-Series Analysis 2.2 XAI - Explainable AI 2.3 Analysing Data Produced by Smart Home Appliances 2.4 Mining Time-Boxed Usage Patterns 3 Proposed Processing Pipeline for Inference of Time-Dependent Usage Pattern 4 Experiments and Results 4.1 Data Description 4.2 Planted Patterns 4.3 Data Preparation and Feature Extraction 4.4 Determing the Existing Usage Patterns with Clustering Algorithms 4.5 Inferring Conditions of Occurrence for the Identified Usage Patterns 5 Conclusions and Future Work References A Deep Learning-Based Technique to Determine Various Stages of Alzheimer’s Disease from 3D Brain MRI Images 1 Introduction 2 Related Work 3 Dataset Overview 4 Proposed Method 5 Image Preprocessing 6 Result Analysis 6.1 Results of Proposed Model 6.2 Comparison 7 Conclusion References Churn Prediction in Enterprises with High Customer Turnover 1 Introduction 2 Foundations 2.1 High Customer Turnover 2.2 Literature Review 3 The CP-HCT Framework 3.1 Data Understanding 3.2 Data Preparation 3.3 Quick Test 3.4 Hyperparameter Setting 3.5 Modeling 3.6 Evaluation 4 Experiments 4.1 Datasets 4.2 Results 5 Conclusions and Future Work References Data Fusion Performance Prophecy: A Random Forest Revelation 1 Introduction 2 Random Forest-Based Fusion Performance Prediction 3 Experimental Settings and Results 4 Conclusions References Celestial Machine Learning 1 Introduction 2 Background and Related Work 3 Methodology 4 Performance Evaluation 5 Conclusion References Deep Learning Based Emotion Recognition Using EEG Signal 1 Introduction 2 Related Work 3 Dataset and Methodology 4 Experiment and Result Analysis 5 Concluding Remarks References Generative AI Generating Fine-Grained Aspect Names from Movie Review Sentences Using Generative Language Model 1 Introduction 2 Related Work 2.1 Language Model in Online Review 2.2 Aspect Extraction 2.3 Text Summarization Using Language Model 3 Aspect Name Generation with T5 3.1 Creating Training Data Using Crowdsourcing 3.2 Additional Pre-training: Predicting Masked Term in Movie Review Data 3.3 1st Intermediate Fine-Tuning: Binary Classification of Whether a Sentence Includes an Aspect 3.4 2nd Intermediate Fine-Tuning: Binary Classification of Whether Two Reviews Have the Same Aspect 3.5 Final Fine-Tuning: Aspect Name Generation 4 Evaluation 4.1 Dataset 4.2 Comparison Methods 4.3 Implementation 4.4 Preliminary Experiment: Automated Accuracy Evaluation 4.5 Experiment: Subject Evaluation of Generated Aspect Names 4.6 Result 5 Discussion 6 Conclusion References Reducing Human Effort in Keyphrase-Based Human-in-the-Loop Topic Models: A Method for Keyphrase Recommendations 1 Introduction 2 Human-in-the-loop Topic Modeling 3 Keyphrase Recommendations 4 Evaluation Experiments 4.1 Experimental Setup 4.2 Empirical Findings and Interpretation 4.3 Discussion 5 Conclusion References Movie Keyword Search Using Large-Scale Language Model with User-Generated Rankings and Reviews 1 Introduction 2 Related Work 3 Movie Keyword Search Using Review and LLM 3.1 Vectorizing Movies Using Reviews 3.2 Formatting User-Generated Rankings as Training Data 3.3 Learning the Relationship Between Movies and Words in the Ranking Title 4 Evaluation 4.1 Experiment 4.2 Experimental Results 5 Discussion 6 Conclusion References Extraction News Components Methods for Fake News Correction 1 Introduction 2 Related Work 3 Definition of Fake News 4 Extraction of Fake Passage 4.1 News Component 4.2 Proposed Methods 5 Experiments 5.1 Evaluation of the Proposed Model 5.2 Comparison of Proposed Methods with the Baseline 6 Conclusion References Image Data and Knowledge Graph Buy Eye-Mask Instead of Alarm Clock!: Graph-Based Approach to Identify Functionally Equal Alternative Products 1 Introduction 2 Related Work 2.1 Recommendation and Search of Products by Purpose 2.2 Serendipity of Recommendation 2.3 Achievement Products for the Purpose 3 Graph-Based Method to Identify Alternative Products 3.1 Extracting Purposes from Product Review Data 3.2 Creating Nodes Comprising Multiple Purpose Terms 3.3 Creating Bipartite Graph of Product and Purpose 3.4 Edge Weighting 3.5 Random Walk with Restart 4 Evaluation 4.1 Dataset 4.2 Implementation 4.3 Compared Method 4.4 Evaluation of Alternative Products 4.5 Evaluation of Ranking 4.6 Experimental Results 5 Discussion 6 Conclusion References Deep Image Analysis for Microalgae Identification 1 Introduction 2 Materials and Methods 3 Results 4 Discussion 5 Conclusions References Provenance-Aware Data Integration and Summarization Querying for Knowledge Graphs 1 Introduction 1.1 Related Work 2 Preliminaries and Problem Statement 3 The Proposed Approach 3.1 Provenance-Aware Data Integration 3.2 Provenance-Aware Summarization Querying 4 Implementation and Experimental Results 4.1 The Data Sets 4.2 The Implementation and Experimental Setup 4.3 The Experimental Results 5 Conclusion References Integration of Knowledge Bases and External Information Sources via Magic Properties and Query-Driven Entity Linking 1 Introduction 2 Related Work 3 Preliminaries 4 Knowledge Mediator 4.1 Extended Knowledge Graph 4.2 Overall Architecture 4.3 Registration of ESPs 4.4 Mapping External Objects to KB Entities 4.5 Execution of Extended SPARQL Query 5 Entity Linking (EL) 5.1 Candidate Selection (CS) 5.2 Linking Decision (LD) 6 Experiment 6.1 Experimental Environment 6.2 Experimental Results 7 Conclusion References Enhancing Taxi Placement in Urban Areas Using Dominating Set Algorithm with Node and Edge Weights 1 Introduction 1.1 Motivation and Contributions 2 Literature Review 3 Preliminaries 4 Problem Definition 5 Proposed Approach 5.1 Dominating Set Example 5.2 Node Entropy 5.3 Weighted Degree 5.4 Ensembled Node Entropy and Weighted Degree with k-Hop Dominating Set 5.5 Recommendation Phase 6 Experiments and Results 6.1 Experimental Setup 6.2 Result Analysis 7 Conclusion and Future Scope References Fast Correlated DNA Subsequence Search via Graph-Based Representation 1 Introduction 2 Preliminary 3 Proposed Method: Pruning-Based Algorithm 3.1 A DNA Graph 3.2 Graph-Based Pruning Rules 3.3 Algorithm 4 Experiments 4.1 Experimental Setup 4.2 Efficiency 4.3 Accuracy 5 Conclusion References Efficient Maximum k-plex Search via Selective Branch-and-Bound 1 Introduction 1.1 Existing Approaches and Challenges 1.2 Our Approaches and Contributions 2 Preliminaries 2.1 Maximum k-plex Search (MPS) Problem 2.2 Branch-and-Bound (BnB) 3 Selective Branch-and-Bound (SBnB) 3.1 Ideas 3.2 Removable Nodes 3.3 Selective Graph Reduction 3.4 Algorithm, Optimization, and Analysis 4 Experimental Evaluation 4.1 Efficiency 4.2 Effectiveness of Selective Graph Reduction 5 Conclusion References Automatic Hypotheses Testing Over Heterogeneous Biological Databases Using Open Knowledge Networks 1 Introduction 2 Proposed Architecture of BioNursery 3 Crowd Computing as an Open Knowledge Ecosystem 4 Conclusion and Future Research References Recommendation Systems Using Derived Sequential Pattern Mining for E-Commerce Recommendations in Multiple Sources 1 Introduction 2 Related Work 3 The Proposed Multi-source Recommendation System MCE-HSPRec 3.1 Running Example of MCE-HSPRec (Algorithms 1–5) 4 Results and Analysis 5 Conclusion and Future Work References Taste Representation Learning Toward Food Recommendation Balancing Curiosity and Comfort 1 Introduction 2 Related Work 3 Proposed Method 3.1 JNEM 3.2 Component Description 3.3 Estimation Method for Ingredient and Taste Representation Vectors 3.4 Training Task of Taste Network 3.5 Dataset 3.6 Learning and Evaluation of the Taste Network 4 User Experiment and Case Study 4.1 User Experiment 4.2 Case Study 5 Conclusion References Tag2Seq: Enhancing Session-Based Recommender Systems with Tag-Based LSTM 1 Introduction 2 Related Work 3 Tag2Seq Recommendation Model 3.1 Embedding Layers 3.2 LSTM 4 Experimental Evaluation 4.1 Dataset Description 4.2 Results and Discussion 5 Conclusion References ClinLearning: An Online Clinical Tutoring and Crowdsourced Treatment Recommendation System 1 Introduction 2 A Document Model for Cases 2.1 Medical Knowledge 3 Components of ClinLearning 4 Conclusion and Future Research References Similarity Measure and Metric The Actualization of TikTok Affordances to Challenge Female Unrealistic Standards of Beauty 1 Introduction 2 Literature Review 2.1 TikTok as a Social Media Application 2.2 Social Media Followers and Influencers 2.3 Social Media and Societal Standards of Beauty 2.4 Self-Presentation on Social Media 2.5 Impression Formation on Social Media 3 Theoretical Framework 3.1 Affordance Theoretical Framework 3.2 Social Media Affordances 4 Methodology 5 Findings 5.1 Association 5.2 Editability 5.3 Sharing 5.4 Browsing Others’ Content 6 Discussion 7 Conclusion References Novel Blocking Techniques and Distance Metrics for Record Linkage 1 Introduction 2 A Summary of Blocking Techniques 2.1 Our Novel Blocking Techniques 3 Distance Metrics 4 Novel Distance Metrics 5 Experimental Results 5.1 Experimental Setup 5.2 Experimental Data 5.3 Results for Distance Metrics 5.4 Results for Blocking 6 Conclusions References Efficiently Discovering Spatial Prevalent Co-location Patterns Without Distance Thresholds 1 Introduction 2 Related Work 3 Method 3.1 Basic Concepts 3.2 Determining Neighboring Instances by Delaunay Triangulation 4 DT k-Order Clique-Based Mining SPCP Algorithm 4.1 Overview of the Mining Algorithms 4.2 Depth-First Search Clique Instance Drive Schema 4.3 Compressed Clique Hash Table 4.4 The DTkC Algorithm 5 Experimental Results and Analysis 5.1 Experiment Setting 5.2 Compare the Mining Performance 5.3 Evaluate the Scalability of DTkC 6 Conclusion References Boosting Similar Compounds Searches via Correlated Subgraph Analysis 1 Introduction 1.1 Existing Approaches and Challenges 1.2 Our Approach and Contributions 2 Preliminary 3 Proposed Method: Correlation-Aware Approach 3.1 Basic Ideas 3.2 Algorithm 4 Experimental Analysis 4.1 Accuracy 4.2 Efficiency 4.3 Impacts of Parameters 5 Conclusion References Efficient Similarity Searches for Multivariate Time Series: A Hash-Based Approach 1 Introduction 1.1 Existing Approaches and Challenges 1.2 Our Approaches and Contributions 2 Preliminary 3 Proposed Method: A Hash-Based Approach 3.1 Basic Concept 3.2 (Step 1) LSH-Based Subsequence Projection 3.3 (Step 2) Subsequence Grouping 3.4 (Step 3) LSH-Based Query Processing 4 Experimental Analysis 4.1 Efficiency 4.2 Scalability 4.3 Accuracy 5 Conclusion References VIII Topic and Text Matching A Machine Learning Approach to Enterprise Matchmaking Using Multilabel Text Classification Based on Semi-structured Website Content 1 Introduction 2 State of the Art and Related Works 2.1 Enterprise Matchmaking 2.2 Text Classification 3 Generic Matchmaking Approach 4 Case Study 4.1 Defining Domain Characteristics 4.2 Initial Dataset Creation 4.3 Mapping Domain Characteristics to Categories 4.4 TARS Specific Data Set Optimization 4.5 Model Parametrization 5 Results 6 Conclusion References TraPM: A Framework for Online Pattern Matching Over Trajectory Streams 1 Introduction 2 Related Work 2.1 Event Pattern Matching 2.2 Pattern Matching over Trajectories 2.3 Distributed Stream Processing 3 Preliminaries and Problem Definition 3.1 Preliminaries 3.2 Problem Definition 4 TraPM: A Framework for Online Pattern Matching over Trajectory Streams 4.1 Proposed Architecture 4.2 Query Language 4.3 Pattern Matching Evaluation 4.4 Scalable Pattern Matching over Trajectory Streams 5 Experimental Evaluation 5.1 Dataset 5.2 Experimental Setting 5.3 Experimental Result 6 Conclusion and Future Work References Attention Based Stopword Generation for Neural Network Based Text Processing 1 Introduction 2 Related Work 3 Proposed Method 3.1 Attention 3.2 Probabilistic Stopword Removal 4 Experiments 4.1 Procedure 4.2 Experiment 1. Rakuten Dataset: Evaluation Point 4.3 Experiment 2. Rakuten Dataset: Usage 4.4 Experiment 3. Twitter Dataset 4.5 Experiment 4. Livedoor News Corpus 5 Conclusions References Digital Index Card Creation and Management for Memorizing What You See on the Web 1 Introduction 2 Related Work 2.1 Web Browsing History Analysis 2.2 Card Making in Education 2.3 Lifelog Analysis 3 Method 3.1 Extraction of Phrase Candidates from Website Browsing History 3.2 Memory Retention Support with Semi-Manual Card Making 3.3 Reflection Support with Card Management Interface 4 Evaluation 4.1 Comparison Methods 4.2 Experimental Tasks 4.3 Experimental Results 5 Discussion 6 Conclusion References Multitask, Cross-Lingual Recipe Classification Using Joint Fine-Tuning Mechanisms 1 Introduction 2 Proposed Solution 2.1 Data Description 2.2 Joint Classification Model 3 Experiments 3.1 Comparison with Baselines 3.2 Feature Ablation Study 3.3 Effect of In-Domain Pre-training 3.4 Cross-Lingual Capabilities 4 Conclusions References Correction to: Integration of Knowledge Bases and External Information Sources via Magic Properties and Query-Driven Entity Linking Correction to: Chapter 30 in: P. Delir Haghighi et al. (Eds.): Information Integration and Web Intelligence, LNCS 14416, https://doi.org/10.1007/978-3-031-48316-5_30 Author Index