Smart Applications with Advanced Machine Learning and Human-Centred Problem Design

General Committees
Honorary Chairs
General Chair
Conference Chairs
Organizing Committee
Secretary and Social Media
Accommodation and Registration/Venue Desk
Travel/Transportation
Web/Design/Conference Session
Scientific Committee
Keynote Speaks
ICAIAME 2021 Keynote Speakers
Foreword
Preface
Contents
About the Conference
Scope/Topics
Conference Scope/Topics (as not limited to)
Conference Posters
1 Implementation of Basic Math Processing Skills with Neural Arithmetic Expressions in One and Two Stage Numbers
1.1 Introduction
1.1.1 Neural Arithmetic Expressions and Logic Units
1.1.2 Long Short-Term Memory Algorithm
1.2 Related Work
1.3 Proposed Method
1.4 Experimental Findings
1.5 Conclusions
References
2 An Example Application for Early Diagnosis of Retinal Diseases Using Deep Learning Methods
2.1 Introduction
2.2 Material and Method
2.2.1 Material
2.2.2 Method
2.3 Research Findings
2.4 Discussion
2.5 Results
References
3 Autonomous Parking with Continuous Reinforcement Learning
3.1 Introduction
3.2 Related Works
3.2.1 Deep Q Networks
3.2.2 Deep Deterministic Policy Gradient Algorithm
3.2.3 Twin Delayed Temporal Difference Algorithm
3.2.4 Soft Actor Critic Algorithm
3.2.5 Hindsight Experience Replay Algorithm
3.2.6 Parking Environment Simulation Model
3.3 Experiments and Results
3.4 Conclusions and Future Work
References
4 Design and Manufacturing of a 3 DOF Robot with Additive Manufacturing Methods
4.1 Introduction
4.2 Material and Method
4.2.1 Material
4.3 Method
4.4 Findings and Discussion
4.5 Conclusion
References
5 Real-Time Mask Detection Based on Artificial Intelligence Using Renewable Energy System Unmanned Aerial Vehicle
5.1 Introduction
5.2 Related Studies
5.3 Material and Method
5.3.1 Material
5.3.2 Material and Method
5.4 Research Findings
5.5 Conclusion
References
6 Investigation of Effect of Wrapping Length on the Flexural Properties of Wooden Material in Reinforcement with Aramid FRP
6.1 Introduction
6.2 Material and Method
6.3 Results
6.4 Conclusions
References
7 Deep Learning-Based Air Defense System for Unmanned Aerial Vehicles
7.1 Introduction
7.2 Material and Method
7.2.1 Material
7.2.2 Method
7.3 Research Findings
7.3.1 MobileNetV2 Training Results
7.3.2 Xception Training Results
7.3.3 InceptionV3 Training Results
7.4 Results
References
8 Strategic Framework for ANFIS and BIM Use on Risk Management at Natural Gas Pipeline Project
8.1 Introductıon
8.2 Literature
8.3 Materials and Methods
8.3.1 Artificial Neural Networks (ANN)
8.3.2 Structure of Artificial Neural Network
8.3.3 Fuzzy Inference System
8.3.4 Adaptive Neuro-Fuzzy Inference System—ANFIS
8.3.5 What is the Building Information Modelling (BIM)
8.3.6 Methods
8.4 Results
8.5 Conclusion
References
9 Predicting Ethereum Price with Machine Learning Algorithms
9.1 Introduction
9.2 Related Works
9.3 Method and Material
9.3.1 Used Methods
9.3.2 Data Collecting
9.3.3 Method
9.4 Discussion and Results
9.5 Conclusions and Future Work
References
10 Data Mining Approachs for Machine Failures: Real Case Study
10.1 Introductıon
10.2 Literature
10.3 Methods
10.3.1 Re-processing the Data
10.3.2 Methods
10.4 Results
10.5 Conclusion
References
11 Classification of People Both Wearing Medical Mask and Safety Helmet
11.1 Introduction
11.2 Materials and Methods
11.2.1 Dataset
11.2.2 Method
11.2.3 Single Deep Neural Network
11.2.4 Double Deep Neural Network
11.3 Conclusions and Future Work
References
12 Anonymization Methods for Privacy-Preserving Data Publishing
12.1 Introduction
12.2 Big Data Definition
12.3 Data Anonymization
12.3.1 Protection Methods with Anonymization
12.3.2 Anonymization and Protection Models
12.4 Literature Review
12.5 Comparison of Existing Studies
12.6 Conclusion
References
13 Improving Accuracy of Document Image Classification Through Soft Voting Ensemble
13.1 Introduction
13.2 Related Works
13.3 Methodology
13.3.1 Document Image Classification
13.3.2 Image Pre-processing
13.3.3 Convolutional Neural Network
13.3.4 Soft Voting
13.4 Experiments and Result
13.4.1 Dataset
13.4.2 Evaluation Metrics
13.4.3 Experiments
13.5 Conclusion
References
14 Improved Performance of Adaptive UKF SLAM with Scaling Parameter
14.1 Introduction
14.2 Adaptive UKF SLAM
14.3 Simulation Results and Discussions
14.4 Conclusion and Suggestions
References
15 An Adaptive EKF Algorithm with Adaptation of Noise Statistic Based on MLE, EM and ICE
15.1 Introduction
15.2 Methods
15.2.1 Extended Kalman Filter (EKF)
15.2.2 Unscented Kalman Filter (UKF)
15.2.3 Adaptive Extended Kalman Filter (AEKF)
15.2.4 Data Association
15.2.5 AEKF-SLAM Algorithm
15.3 Simulation Results and Discussion
15.4 Conclusions and Future Work
References
16 Artificial Intelligence Based Detection of Estrus in Animals Using Pedometer Data
16.1 Introduction
16.2 Related Works
16.3 Method and Material
16.3.1 Architectural Design
16.3.2 Devices
16.3.3 Electronic Circuit Design
16.3.4 Proposed Algorithms
16.4 Discussion and Result
16.5 Conclusions and Future Work
References
17 Enhancing Lexicon Based Sentiment Analysis Using n-gram Approach
17.1 Introduction
17.2 Sentiment Lexicons
17.2.1 Vader
17.2.2 TextBlob
17.2.3 Afinn
17.2.4 SentiWordNet
17.3 Proposed Framework
17.3.1 Pre-processing Step
17.3.2 N-gram Extraction
17.3.3 Feature Space Construction
17.4 Experimental Results
17.5 Conclusion
References
18 A Comparison of Word Embedding Models for Turkish
18.1 Introduction
18.2 Data and Data Preprocessing Steps
18.3 Method
18.3.1 Embedding Models
18.3.2 Classification Model
18.4 Experiments
18.5 Conclusion
References
19 The Unfairness of Collaborative Filtering Algorithms' Bias Towards Blockbuster Items
19.1 Introduction
19.2 Related Works
19.3 Description of Blockbuster Items
19.4 Blockbuster Bias in User Profiles
19.4.1 The Propensities of Users for Blockbuster Items
19.4.2 Profile Size and Blockbuster Bias
19.5 Different User Groups in Terms of Inclination for Blockbuster
19.6 Algorithmic Propagation of Blockbuster Bias
19.6.1 Blockbuster Bias in Recommendations for Different User Groups
19.7 Conclusion and Future Work
References
20 Improved Gradient-Based Optimizer with Dynamic Fitness Distance Balance for Global Optimization Problems
20.1 Introduction
20.2 Related Works
20.2.1 GBO
20.2.2 Dynamic Fitness-Distance Balance (dFDB)
20.2.3 Improved GBO with Dynamic Fitness Distance Balance
20.3 Experimental Study
20.3.1 Settings
20.3.2 Benchmark Problems
20.3.3 Constrained Engineering Design Problems
20.4 Analyze Results
20.4.1 Statistical Analysis Results
20.4.2 Convergence Analysis Results
20.4.3 Results for Engineering Design Problems
20.5 Conclusions and Future Work
References
21 TR-SUM: An Automatic Text Summarization Tool for Turkish
21.1 Introduction
21.2 Literature Review
21.2.1 Related Studies in Turkish
21.2.2 Datasets in Turkish
21.3 TR-SUM: A Text Summarization Tool for Turkish
21.3.1 General Overview of “TR-SUM: A Text Summarization Tool for Turkish”
21.3.2 TR-NEWS-SUM Dataset
21.3.3 Data Pre-processing
21.3.4 The Proposed Neural Network Models for Turkish Text Summarization
21.4 Discussion and Results
21.5 Conclusion and Future Work
References
22 Automatic and Semi-automatic Bladder Volume Detection in Ultrasound Images
22.1 Introduction
22.2 Related Works
22.3 Method and Material
22.3.1 Data Set
22.3.2 Method
22.4 Discussion and Results
22.5 Conclusions and Future Work
References
23 Effects of Variable UAV Speed on Optimization of Travelling Salesman Problem with Drone (TSP-D)
23.1 Introduction
23.2 Problem Definition
23.3 Methodology
23.3.1 Truck-Drone Algorithm Approach
23.4 Experimental Studies
23.4.1 Settings
23.4.2 Experimental Studies and Results
23.5 Discussions and Conclusion
References
24 Improved Phasor Particle Swarm Optimization with Fitness Distance Balance for Optimal Power Flow Problem of Hybrid AC/DC Power Grids
24.1 Introduction
24.2 Mathematical Formulation of Optimal Power Flow Problem of Hybrid AC/DC Power Grids
24.2.1 State and Control Variables
24.2.2 Constraints
24.2.3 Objective Functions
24.3 Method
24.3.1 Fitness-Distance Balance Method
24.3.2 Overview of Phasor Particle Swarm Optimization (PPSO) Algorithm
24.3.3 Proposed FDBPPSO Algorithm
24.4 Experimental Settings
24.5 Results and Analysis
24.5.1 Determining the Best FDBPPSO Variant on CEC 2020 Test Suite
24.5.2 Application of the Proposed FDBPPSO Method for Optimal Power Flow Problem of Hybrid AC/DC Power Grids
24.6 Conclusions
References
25 Development of an FDB-Based Chimp Optimization Algorithm for Global Optimization and Determination of the Power System Stabilizer Parameters
25.1 Introduction
25.2 Mathematical Formulation of Power System Stabilizer Parameters Optimization
25.2.1 Power System Model with PSS Structure
25.2.2 Objective Functions and Constraints
25.3 Method
25.3.1 Fitness-Distance Balance Selection Method
25.4 Overview of Chimp Optimization Algorithm
25.4.1 Proposed FDBChOA Algorithm
25.5 Experimental Settings
25.6 Results and Analysis
25.6.1 Determining the Best FDBPPSO Variant on CEC 2020 Benchmark Test Suite
25.6.2 Application of the Proposed FDB- Based Chimp Optimization Algorithm for Power System Stabilizer Parameters Optimization
25.7 Conclusions
References
26 Deep Learning-Based Prediction Model of Fruit Growth Dynamics in Apple
26.1 Introduction
26.2 Materials and Methods
26.3 Results and Discussion
References
27 Prediction of Hepatitis C Disease with Different Machine Learning and Data Mining Technique
27.1 Introduction
27.2 Materials and Method
27.2.1 Dataset Introduction
27.2.2 Data Mining Process
27.2.3 Machine Learning Methods
27.3 Experimental Results
27.3.1 Evaluation Metrics
27.3.2 Results and Findings
27.4 Conclusions and Future Work
References
28 Prediction of Development Types from Release Notes for Automatic Versioning of OSS Projects
28.1 Introduction
28.2 Related Works
28.3 Method and Material
28.3.1 Dataset
28.3.2 Pre-processes
28.3.3 Methods
28.3.4 Model Evaluation
28.4 Results
28.5 Discussion
References
29 Design Optimization of Induction Motor with FDB-Based Archimedes Optimization Algorithm for High Power Fan and Pump Applications
29.1 Introduction
29.2 Mathematical Formulation of Optimization Problem
29.3 Method
29.3.1 Archimedes Optimization Algorithm
29.3.2 Archimedes Optimization Algorithm (AOA) with Fitness Distance Balance
29.4 Experimental Settings
29.5 Results and Analysis
29.5.1 Determining the Best FDB-AOA Method on Benchmark Problems
29.5.2 Application of the Proposed FDB-AOA Method for Design Optimization of Induction Motor
29.6 Conclusions
References
30 Collecting Health Information with LoRa Technology
30.1 Introduction
30.2 Related Works
30.3 Material and Method
30.3.1 LoRa Communication
30.3.2 Node Part
30.3.3 Server Part
30.3.4 Client Part
30.3.5 Mobile Application
30.3.6 Wearable Module Hardware
30.4 Discussion and Results
30.5 Conclusions and Future Work
References
31 A New Hybrid Method for Indoor Positioning
31.1 Introduction
31.2 Related Works
31.3 Material and Method
31.3.1 System Architecture
31.3.2 Indoor Positioning
31.4 Discussion and Results
31.5 Conclusions and Future Work
References
32 On the Android Malware Detection System Based on Deep Learning
32.1 Introduction
32.1.1 Previous Works
32.1.2 Motivation and Contribution
32.1.3 Organization
32.2 Experimental Settings
32.2.1 Used Datasets
32.2.2 Performance Measure
32.3 Methodologies
32.3.1 Static Analysis
32.3.2 Converting Static Properties to Images
32.3.3 Deep Learning Techniques
32.4 Results and Discussions
32.4.1 Results with Malgenome-215 Dataset
32.4.2 Results with the Drebin-215 Dataset
32.5 Conclusion and Future Works
References
33 Poisson Stability in Inertial Neural Networks
33.1 Introduction
33.2 Main Result
33.3 Numerical Example
References
34 Poisson Stable Dynamics of Hopfield-Type Neural Networks with Generalized Piecewise Constant Argument
34.1 Introduction
34.2 Preliminaries
34.3 Main Result
34.4 Example
34.5 Conclusions
References
35 A Business Workflow for Clustering and Decision Making Systems in Tax Audit Industry: A Case Study
35.1 Introduction
35.2 Literature Review
35.2.1 Fundamental Concepts
35.3 Methodology
35.3.1 Clustering Algorithm Module Utilizing Container Based Virtualization
35.3.2 Rule Based Decision Making Module Utilizing Container Based Virtualization
35.4 Conclusion and Future Work
References
36 Mask R-CNN Approach for Egg Segmentation and Egg Fertility Classification
36.1 Introduction
36.2 Literature Review
36.3 Method and Material
36.4 Implementation
36.5 Results
36.6 Discussion
References
37 Optimizing the Hedging Rules for the Dam Reservoir Operations by Meta-Heuristic Algorithms
37.1 Introduction
37.2 Study Region and Data
37.3 Methodology
37.3.1 Hedging Models Used
37.3.2 Model Calibrations
37.4 Results
37.5 Conclusion
References
38 Next Word Prediction with Deep Learning Models
38.1 Introduction
38.2 Related Works
38.3 Method and Material
38.3.1 Dataset
38.3.2 NLP with Deep Learning
38.3.3 Modeling
38.4 Discussion and Results
38.4.1 RNN-GRU Model
38.4.2 LSTM Model
38.4.3 Human Experiments
38.5 Conclusions and Future Work
References
39 Cooperative Multi-agent Reinforcement Learning for Autonomous Cars Passing on Narrow Road
39.1 Introduction
39.2 Related Works
39.3 Method
39.3.1 Problem Definition
39.3.2 Reward Function
39.3.3 Agent Networks
39.3.4 Curriculum Learning
39.4 Experiments
39.5 Conclusions and Future Work
References
40 Oscillations in Recurrent Neural Networks with Structured and Variable Impulses
40.1 Introduction
40.1.1 The Structure of the Model
40.1.2 Basic Conditions of the Research
40.2 Almost Periodic Solutions
40.3 Periodic Solutions
40.4 Conclusion
References
41 Topic Modeling Analysis of Tweets on the Twitter Hashtags with LDA and Creating a New Dataset
41.1 Introduction
41.2 Literature
41.2.1 The Problems Posed by Tweeting
41.2.2 Studies Conducted with Artificial İntelligence Conducted on Twitter in the Literature
41.2.3 Studies Conducted with Artificial İntelligence Conducted on Twitter in the Literature
41.2.4 The Process of Natural Language Processing
41.3 Material-Method
41.3.1 Data Collection
41.3.2 Data Processing
41.3.3 Creating a DATASET
41.3.4 Analysis and Classification
41.3.5 Topical Modeling
41.4 Research Findings
41.4.1 Bi-gram, Tri-gram
41.5 Conclusion and Suggestions
References
42 Hopfield-Type Neural Networks with Poincaré Chaos
42.1 Introduction
42.2 Preliminaries
42.3 Main Result
42.4 Examples
References
43 Face Expression Recognition Using Deep Learning and Cloud Computing Services
43.1 Introduction
43.2 Related Works
43.3 Method and Material
43.3.1 Deep Learning
43.3.2 Convolutional Neural Networks
43.3.3 Cloud Computing Services
43.4 Results and Discussion
43.5 Conclusions and Future Work
References
44 Common AI-Based Methods Used in Blood Glucose Estimation with PPG Signals
44.1 Introduction
44.2 AI-Based Non-invasive BGL Methods
44.2.1 Pulse Based Cepstral Coefficients
44.2.2 Support Vector Machine (SVM)
44.2.3 Decision Tree (DT)
44.2.4 Random Forest Regression (RFR)
44.2.5 K-Nearest Neighbor (KNN)
44.2.6 Artificial Neural Network (ANN)
44.2.7 Naïve Bayes (NB)
44.3 Conclusion and Suggestions
References
45 Capturing Reward Functions for Autonomous Driving: Smooth Feedbacks, Random Explorations and Explanation-Based Learning
45.1 Introduction
45.2 Problem Formulation
45.3 Method
45.4 Experiments
45.4.1 The Environment
45.4.2 The Interface
45.4.3 Evaluation
45.5 Conclusion
References
46 Unpredictable Solutions of a Scalar Differential Equation with Generalized Piecewise Constant Argument of Retarded and Advanced Type
46.1 Introduction
46.2 Preliminaries
46.3 Results on Unpredictable Solutions
46.4 Example with a Numerical Simulation
46.5 Conclusion
References
47 Classification of Naval Ships with Deep Learning
47.1 Introduction
47.2 Related Works
47.3 Dataset
47.4 Classification
47.5 Conclusions
References
48 Investigation of Mass-Spring Systems Subject to Generalized Piecewise Constant Forces
48.1 Introduction and Preliminaries
48.2 Dynamics of Mass-Spring Systems Subject to Generalized Piecewise Constant Forces
48.2.1 Undamped Spring-Mass System
48.2.2 Damped Spring-Mass System
48.3 Conclusion
References
49 Classification of High Resolution Melting Curves Using Recurrence Quantification Analysis and Data Mining Algorithms
49.1 Introduction
49.2 Materials and Methods
49.2.1 Dataset
49.2.2 Methods
49.2.3 Proposed Method
49.3 Experimental Results
49.4 Conclusions
References
50 Machine Learning Based Cigarette Butt Detection Using YOLO Framework
50.1 Introduction
50.2 Related Works
50.3 Method and Material
50.3.1 Deep Learning
50.3.2 Convolutional Neural Network (CNN)
50.3.3 You Only Look Once (YOLO)
50.3.4 Dataset
50.4 Results and Discussion
50.5 Conclusions and Future Work
References
51 Securing and Processing Biometric Data with Homomorphic Encryption for Cloud Computing
51.1 Introduction
51.2 Related Works
51.3 Method and Material
51.3.1 Biometric Identification
51.3.2 Homomorphic Encryption
51.3.3 Overview of SEAL and TENSEAL
51.4 Proposed Methodology and Algorithm
51.4.1 Experimental Dataset
51.5 Discussion and Results
51.6 Conclusions and Future Work
References
52 Automatic Transferring Data from the Signed Attendance Papers to the Digital Spreadsheets
52.1 Introduction
52.2 Computer Vision Methods
52.2.1 Canny Edge Detection
52.2.2 Morphological Tranformations
52.2.3 Shape Skeleton
52.3 Convolutional Neural Networks
52.3.1 Convolution Layer
52.3.2 Rectified Linear Unit (ReLU) Layer
52.3.3 Max-Pooling Layer
52.3.4 Fully Connected Layer
52.4 Proposed Methods
52.5 Results
52.6 Conclusion
References
53 Boarding Pattern Classification with Time Series Clustering
53.1 Introduction
53.2 Methodology
53.3 Results and Discussion
53.4 Conclusion
References
54 Shipment Consolidation Practice Using Matlog and Large-Scale Data Sets
54.1 Introduction
54.2 Background
54.3 Shipment Consolidation Problem
54.3.1 TL Transport Charge
54.3.2 Total Logistics Cost
54.4 Methodology
54.5 Computational Experiments
54.5.1 The Second Phase: Determination of Consolidated Shipments and the Shipment Routes
54.5.2 Computational Experiments with Large-Scale Data Sets
54.6 Conclusion
54.7 Appendix 1 Some of the Solution Graphics for Aegean Town Data Sets
54.8 Appendix 2 Some of the Solution Graphics for Turkey Data Sets
References
55 The Imminent but Slow Revolution of Artificial Intelligence in Soft Sciences: Focus on Management Science
55.1 Introduction
55.2 Background and Related Works
55.2.1 Artificial Intelligence
55.2.2 Soft Sciences
55.3 Problem Position and Research Gap
55.4 Proposed Approach and Methodology
55.4.1 Investigating the Contrasted Investments on AI Research in Management Science: Visual Analysis
55.4.2 Investigating AI Research in the Sub-fields of Management Science: Quantitative Analysis
55.4.3 Investigating AI Impacts on Management Research: Qualitative Analysis
55.5 Discussion and Outcomes
55.6 Conclusion, Limitations, and Perspectives
References
56 Multi-criteria Decision-Making for Supplier Selection Using Performance Metrics and AHP Software. A Literature Review
56.1 Introduction
56.2 Methodology
56.2.1 Literature Review
56.3 Results
56.3.1 Criteria for Selecting a Supplier
56.3.2 Tools for the Selecting Process
56.4 Conclusions
References
57 PID Controller and Intelligent Control for Renewable Energy Systems
57.1 Introduction
57.2 Brief History of PID Controllers and Their Operation
57.3 Application of the PID Controller
57.4 Importance of Renewable Energy Systems
57.4.1 Methods of Obtaining Renewable Energies
57.5 How Can We Link PID Controllers with Renewable Energy Systems? and How Would This Benefit Us?
57.5.1 Solar Energy
57.5.2 Hydroelectric Power
57.5.3 Wind Energy
57.6 Conclusion and Suggestions
References
58 Machine Learning Applications in the Supply Chain, a Literature Review
58.1 Introduction
58.2 Background
58.3 Methodology
58.4 Review Findings
58.5 Conclusions
References
59 Machine Learning Applications for Demand Driven in Supply Chain: Literature Review
59.1 Introduction
59.2 Literature Review (LR)
59.2.1 Problem
59.2.2 Demand Driven and Its Role in the Supply Chain and Operations Management
59.2.3 Machine Learning (ML), Tools, Techniques, and Technologies
59.2.4 Implementation and Results of Machine Learning Cases and Applications
59.3 Methodology for the Literature Review
59.3.1 LR Searching Phase 1
59.3.2 LR Selecting Phase 2
59.3.3 LR Analyzing Phase 3
59.4 Conclusions and Future Research
59.4.1 Declaration of Competing Interests
References
60 Dynamic Data-Driven Failure Mode Effects Analysis (FMEA) and Fault Prediction with Real-Time Condition Monitoring in Manufacturing 4.0
60.1 Introduction
60.2 Related Work
60.3 Method and Material
60.3.1 Failure Mode Effects Analysis (FMEA) Method
60.3.2 Programmable Logic Controller (PLC)
60.3.3 Kitchen Equipments Manufacturing Company
60.3.4 System Integration: PLC, ERP, C# with WinProLadder
60.4 Discussion and Results
60.5 Conclusions and Future Work
References