Knowledge-Based Software Engineering: 2022: Proceedings of the 14th International Joint Conference on Knowledge-Based Software Engineering

Description

This book contains extended versions of  the works and new research results presented at the 14th International Joint Conference on Knowledge-based Software Engineering (JCKBSE2022). JCKBSE2022 was originally planned to take place in Larnaca, Cyprus. Unfortunately, the COVID-19 pandemic forced it to be rescheduled as an online conference.   JCKBSE is a well-established international biennial conference that focuses on the applications of Artificial Intelligence on Software Engineering. The 14th International Joint Conference on Knowledge-based Software Engineering (JCKBSE2022) was organized by the Department of Informatics of the University of Piraeus, Greece.   This book is a valuable resource for experts and researchers in the field of (knowledge-based) software engineering. It is also valuable to general readers in the fields of artificial and computational intelligence and, more generally, computer science, wishing to learn more about the exciting field of (knowledge-based) software engineering and its applications. An extensive list of bibliographic references at the end of each chapter helps readers to probe deeper into the application areas of interest to them. Preface Artificial Intelligence as Dual Use Technology Attacks in Android Mobile User Interfaces Exposing User Privacy and Personal Data Contents Part I Software Development Techniques and Tools 1 Proposal of a Middleware to Support Development of IoT Firmware Analysis Tools 1.1 Introduction 1.1.1 Background 1.1.2 Contribution 1.2 Characterization of Firmware Vulnerability Detection Methods 1.2.1 Categorization of Features 1.2.2 Research on Firmware Vulnerability Analysis 1.2.3 Result 1.3 Our Approach 1.3.1 Firmware Splitting 1.3.2 Static Strings 1.3.3 Control Flow Graph 1.3.4 Identifying Network Functions 1.4 Conclusions and Future Work References 2 Feature-Based Cloud Provisioning for Rehosting 2.1 Introduction 2.2 Motivating Examples 2.2.1 Trial-and-Error Searches for Optimal Cloud Service Configuration in Design Process 2.2.2 Manual Provisioning with Console Used by Engineers in Construction Process 2.3 Feature-Based Cloud Provisioning Method for Rehosting 2.3.1 Overview of Proposed Method 2.3.2 Cloud Feature Model 2.3.3 Cloud Provisioning Tool 2.4 Evaluation 2.4.1 Evaluation Method 2.4.2 Results 2.5 Discussion 2.5.1 Effects of Application to Design Process 2.5.2 Effects of Application to Construction Process 2.6 Related Work 2.7 Conclusion References 3 Pattern to Improve Reusability of Numerical Simulation 3.1 Introduction 3.2 Background 3.2.1 Modelica Language 3.2.2 Adopting Design Patterns to Modelica 3.3 Patterns for Physical Modeling and Their Adaptation 3.4 Case Study 3.4.1 Moving Ball 3.4.2 SIR Model 3.5 Conclusions References 4 SpiderTailed: A Tool for Detecting Presentation Failures Using Screenshots and DOM Extraction 4.1 Introduction 4.2 Proposed Method 4.2.1 Take Screenshots 4.2.2 Extract Visual Properties 4.2.3 Comparison of Visual Properties 4.2.4 Output of Comparison Results 4.3 Implementation 4.3.1 Take Screenshots Function 4.3.2 Extract Visual Properties Function 4.3.3 Comparison of Visual Properties Function 4.3.4 Comparison Results Output Function 4.4 Evaluation 4.4.1 Preparing Web Page for Evaluation 4.4.2 Establishment of Evaluation Criteria 4.4.3 Experiment: 1 Evaluation of SpiderTailed 4.4.4 Experiment: 2 Evaluation of Manual Observation 4.5 Results and Discussion 4.5.1 RQ1: What Are the Strengths of SpiderTailed Compared to the Manual Observation 4.5.2 RQ2: Comparison of the Time Consumption 4.5.3 RQ3: Challenges of the SpiderTailed Method 4.6 Related Work 4.7 Limitations and Validity 4.7.1 Limitations 4.7.2 Validity 4.8 Conclusion References Part II AI/ML-Based Software Development 5 Collecting Insights and Developing Patterns for Machine Learning Projects Based on Project Practices 5.1 Introduction 5.2 Related Work 5.3 Research Subject and Hypothesis 5.3.1 ML-based Service System 5.3.2 Architecture Design Pattern for ML Service Systems 5.3.3 Research Hypothesis 5.4 Proposed Method 5.4.1 Overview 5.4.2 Reference Development Model and Collection of Insights 5.4.3 Construction of Patterns from Collected Insights 5.5 Practice 5.6 Discussion 5.7 Conclusions References 6 Supporting Code Review by a Neural Network Using Program Images 6.1 Introduction 6.2 Related Work 6.3 CNN-BI System 6.3.1 Training Method and the Training Data 6.3.2 Preparing the Learning Data 6.3.3 Check List 6.4 Experimental Study 6.4.1 Overview 6.4.2 Applying Supervised Learning 6.4.3 Visualization of the Training 6.4.4 Verification of the Categorization 6.4.5 Types of Defects Inferred 6.4.6 Review Process 6.4.7 Result 6.5 Discussion 6.5.1 The Answer to RQ1 6.5.2 The Answer to RQ2 6.5.3 Internal Validity 6.5.4 External Validity 6.6 Conclusion References 7 Safety and Risk Analysis and Evaluation Methods for DNN Systems in Automated Driving 7.1 Introduction 7.2 Related Work 7.2.1 Machine Learning Systems Engineering and Safety 7.2.2 Safety Guidelines and Research Trends for Automated Driving 7.2.3 Model and STAMP and Related Methods 7.3 Safety Challenges for Machine Learning Systems 7.3.1 Safety Challenges of Automated Driving 7.3.2 The “Question” that Forms the Core of the Research 7.3.3 Research Goals 7.4 Proposal of Safety and Risk Analysis and Evaluation Methods for DNN Systems 7.4.1 Step 1: System-level Safety Analysis 7.4.2 Step 2: Scenario and Training Data Generation for High-Risk Scenes 7.4.3 Step 3: DNN Design Modeling and Problem Analysis 7.4.4 Step 4: Design Labels with Safety in Mind 7.4.5 Step 5: Model Evaluation by Risk 7.4.6 Step 6: Setting Evaluation Criteria 7.4.7 Step 7: Model Improvement Through Debugging and Modification Techniques 7.5 Safety Arguments and Case Studies 7.5.1 Safety Arguments for DNN Safety Analysis and Assessment Methodology 7.5.2 Embodiment of Steps 1–3 7.6 Conclusion References 8 Regulation and Validation Challenges in Artificial Intelligence-Empowered Healthcare Applications—The Case of Blood-Retrieved Biomarkers 8.1 Introduction 8.2 Key Issues and Challenges 8.2.1 Biomarkers 8.2.2 Automating Interventions and Patient's Journey 8.2.3 The Importance of Regulation 8.2.4 Validation of Neural Networks in Health Applications and Continuous Integration 8.2.5 The Role of Machine Learning in Health Care 8.3 Related Work 8.4 Building a Blood Exam-Based Personalised Recommender System 8.4.1 Development Methodologies 8.5 Conclusion and Research Key Findings References Part III Educational and Assistive Software 9 Multi-agent Simulation for Risk Prediction in Student Projects with Real Clients 9.1 Introduction 9.2 Software Development Project Course with Real Clients 9.3 Multi-agent Model of Student Projects 9.3.1 NetLogo 9.3.2 Student Project Model 9.3.3 Dependencies Among Tasks in a Project 9.3.4 Task Allocation to Project Members 9.3.5 Members’ Skill and Performance 9.3.6 Risk Prediction by Simulating in Our Model 9.4 Simulation Results 9.5 Questionnaire Survey 9.6 Related Work 9.7 Conclusion and Future Works References 10 Automatic Scoring in Programming Examinations for Beginners 10.1 Introduction 10.2 Preliminaries 10.2.1 Presburger Arithmetic 10.2.2 Notation 10.3 Proposed Methods 10.3.1 Programming Language 10.3.2 Program Verification 10.3.3 Automatic Scoring 10.3.4 Examination System 10.4 Experiments 10.4.1 Count 10.4.2 Bubble Sort 10.4.3 Binary Search 10.5 Discussion 10.6 Conclusion References 11 A Study on Analyzing Learner Behaviors in State Machine Modeling Using Process Mining and Statistical Test 11.1 Introduction 11.2 Theoretical Background 11.2.1 State Machine Model 11.2.2 Model Log 11.2.3 Event Log 11.2.4 Process Mining 11.2.5 Identification of Different Activity and Transition 11.3 Method 11.3.1 Activity and Transition Extraction 11.3.2 Difference Identification 11.4 Experiment 11.4.1 Modeling Task and Answers 11.4.2 Result 11.4.3 Consideration 11.5 Related Work 11.6 Conclusion References 12 Supporting Conveyance of Webpages by Highlighting Text for Visually Impaired Persons 12.1 Introduction 12.2 Related Work 12.3 Emphasis Expressions in this Research 12.4 Outline of Our Method 12.4.1 Approach of Our Method 12.4.2 Structure of Our Method 12.5 Deciding the Reading Voice for Emphasized Expressions 12.5.1 Weighting of Text 12.5.2 Weighting of Voice 12.5.3 Determination of the Reading Method 12.6 Evaluation 12.6.1 Experimental Design 12.6.2 Results 12.6.3 Discussion 12.7 Conclusion References Part IV Requirements Analysis and Software Modeling 13 Comparative Study on Functional Resonance Matrices 13.1 Introduction 13.2 Related Work 13.2.1 Fram 13.2.2 FRAM Matrix Representation 13.2.3 Matrix Representation 13.3 Functional Aspect Resonance Matrix 13.4 Comparative Study on Matrix Representations 13.5 Discussion 13.5.1 Novelty 13.5.2 Effectiveness 13.5.3 Computational Cost 13.5.4 Limitations 13.6 Summary References 14 A Method for Matching Patterns Based on Event Semantics with Requirements 14.1 Introduction 14.2 Related Work 14.3 Proposed Method 14.3.1 Characteristic Semantic Representation 14.4 Experimental Pattern Matching 14.5 Conclusions References 15 Digital SDGs Framework Towards Knowledge Integration 15.1 Introduction 15.2 Related Work 15.2.1 SDGs 15.2.2 Dx 15.2.3 Knowledge Integration 15.3 Issues 15.4 DSDG Framework 15.4.1 Classification of SDGs in Enterprises 15.4.2 DSDG Strategy Map 15.4.3 DSDG Framework 15.4.4 SDGsVCM 15.5 Case Study 15.5.1 DSDG Strategy Map 15.5.2 DSDG Framework 15.5.3 SDGsVCM 15.6 Discussion 15.7 Summary References 16 Hierarchical User Review Clustering Based on Multiple Sub-goal Generation 16.1 Introduction 16.2 Relevant Work 16.3 The Existing Clustering Method 16.3.1 Ward Method 16.4 Comprehensive Clustering Method 16.4.1 LDA Topic Model 16.4.2 The Distance-Based Clustering Algorithm 16.5 Experiment and Evaluation 16.5.1 Purpose of Experiments 16.5.2 Experiment and Discussion 16.6 Conclusion References