Applied Intelligent Decision Making in Machine Learning

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Notes on the Editors and Contributors
Chapter 1: Data Stream Mining for Big Data
	1.1 Introduction
	1.2 Research Issues in Data Stream Mining
	1.3 Filtering and Counting in a Data Stream
		1.3.1 Bloom Filters
		1.3.2 Counting the Frequency of Items in a Stream
		1.3.3 Count Unique Items in a Data Stream
	1.4 Sampling from Data Streams
	1.5 Concept Drift Detection in Data Streams
		1.5.1 Causes of Concept Drift
		1.5.2 Handling Concept Drift
			1.5.2.1 CUSUM Algorithm
			1.5.2.2 The Higia Algorithm
			1.5.2.3 Dynamic Weighted Majority Algorithm
	1.6 Discussion
	References
Chapter 2: Decoding Common Machine Learning MethodsAgricultural Application Case Studies Using Open Source Software
	2.1 Introduction
	2.2 Literature Review
	2.3 Materials and Methods
		2.3.1 Overall ML Model Development Process
		2.3.2 Data Collection
			2.3.2.1 Iris Dataset
			2.3.2.2 Soybean Aphid Dataset
			2.3.2.3 Weed Species Dataset
		2.3.3 Shape Features Extraction
		2.3.4 Data Cleaning
		2.3.5 Feature Selection
			2.3.5.1 Filter Methods
			2.3.5.2 Wrapper Methods
			2.3.5.3 Embedded Methods
			2.3.5.4 Relief Algorithms
		2.3.6 Data Splitting
		2.3.7 The ML Methods
			2.3.7.1 Linear Discriminant Analysis
			2.3.7.2 k-Nearest Neighbor
		2.3.8 Evaluation of ML Methods
			2.3.8.1 Confusion Matrix
			2.3.8.2 Accuracy
			2.3.8.3 Precision
			2.3.8.4 Recall
			2.3.8.5 F-score
	2.4 Results and Discussion
		2.4.1 Results of Evaluated Features from the Dataset
		2.4.2 Selected Features from the Dataset
		2.4.3 Dataset Test of Normality for Model Selection
		2.4.4 Soybean Aphid Identification
			2.4.4.1 Features Ranking
			2.4.4.2 The LDA Model and Evaluation
		2.4.5 Weed Species Classification
			2.4.5.1 Features Ranking
			2.4.5.2 The kNN Model and Evaluation
		2.4.6 Comparison of Results with the Standard Iris Data
	2.5 Conclusions
	Acknowledgments
	References
Chapter 3: A Multi-Stage Hybrid Model for Odia Compound Character Recognition
	3.1 Introduction
	3.2 Background
		3.2.1 General OCR Stages
		3.2.2 Structural Similarity
		3.2.3 Projection Profile and Kendall Rank Correlation Coefficient Matching
		3.2.4 Local Frequency Descriptor
		3.2.5 General Regression Neural Network (GRNN)
	3.3 Proposed Method
	3.4 Experiments
		3.4.1 Dataset Creation
		3.4.2 Experimental Setup
	3.5 Results and Discussion
	3.6 Conclusion and Future Scope
	References
Chapter 4: Development of Hybrid Computational Approaches for Landslide Susceptibility Mapping Using Remotely Sensed Data in East Sikkim, India
	4.1 Introduction
	4.2 Study Materials and Methodology
		4.2.1 Area of Research Study
		4.2.2 Multi-colinearity Assessment (MCT)
		4.2.3 Affecting Factors
		4.2.4 Landslide Inventory Map (LIM)
		4.2.5 Methodology
			4.2.5.1 Hybrid Biogeography-Based Optimization
			4.2.5.2 Hybridization with Differential Evolution
				4.2.5.2.1 The DE/BBO Algorithm
				4.2.5.2.2 Local-DE/BBO
				4.2.5.2.3 Self-Adaptive DE/BBO
		4.2.6 Validation of Models
		4.2.7 Shortly Structured Methodology
	4.3 Results and Discussion
		4.3.1 Importance of the Conditioning Factors on the Occurrences of Landslides
		4.3.2 Application of Hybrid Biogeography-Based Optimization for Landslide Susceptibility Assessment
	4.4 Conclusion
	References
Chapter 5: Domain-Specific Journal Recommendation Using a Feed Forward Neural Network
	5.1 Introduction
	5.2 Literature Survey
	5.3 Content-Based Recommendation System for Domain-Specific Papers
		5.3.1 Scraping and Data Integration (Challenges and Solutions for Data Collection)
			5.3.1.1 Limitations on the Size of the Query Results
				5.3.1.1.1 Fixed Limits
				5.3.1.1.2 Pagination
			5.3.1.2 Dynamic Contents
			5.3.1.3 Access Limitations
				5.3.1.3.1 Masked URL Parameters
				5.3.1.3.2 Robot Recognition and Reverse Turing Tests
				5.3.1.3.3 Changing the Content of Request Headers
				5.3.1.3.4 Selecting Appropriate Cookie Settings
				5.3.1.3.5 Requests and Different Time Intervals
				5.3.1.3.6 Altering the IP Address
		5.3.2 Data Curation
			5.3.2.1 The Complexity of the Integration Operation
		5.3.3 Phase 1: Identifying Candidate Journals
		5.3.4 Phase 2: Ranking Candidate Journals
	5.4 Experimental Results and Discussions
		5.4.1 Configurations
		5.4.2 Result Analysis
	5.5 Conclusion and Future Work
	References
Chapter 6: Forecasting Air Quality in India through an Ensemble Clustering Technique
	6.1 Introduction
	6.2 Related Works
		6.2.1 Air Quality Prediction
		6.2.2 Ensemble Modeling
			6.2.2.1 Variants of Ensemble Models
		6.2.3 Ensemble Clustering
	6.3 Dataset Descriptions
	6.4 Methodology
		6.4.1 Final Cluster Labeling
		6.4.2 METIS Function
			6.4.2.1 METIS Algorithm
		6.4.3 Phases
		6.4.4 Advantages
	6.5 Experimental Results
		6.5.1 Silhouette Coefficient
		6.5.2 Calinski-Harabasz Index
		6.5.3 Davies-Bouldin Index
	6.6 Conclusion
	References
Chapter 7: An Intelligence-Based Health Biomarker Identification System Using Microarray Analysis
	7.1 Introduction
	7.2 Existing Knowledge
	7.3 Classification Model
	7.4 Approaches for Feature Selection
		7.4.1 Shuffled Frog-Leaping Algorithm and Particle Swarm Optimization (SFLA-PSO)
		7.4.2 The Advantage of SFLA
		7.4.3 Algorithm for BSFLA-PSO
	7.5 Experimental Result Analysis
		7.5.1 Dataset Considered for This Experiment
		7.5.2 Normalization
		7.5.3 Details of Classifiers Used in This Experimental Study and Evaluation Metrics
		7.5.4 Result Analysis
			7.5.4.1 Performance of Proposed BSFLA-PSO with Prostate Dataset
			7.5.4.2 Performance of Proposed BSFLA-PSO with Leukemia Dataset
			7.5.4.3 Performance of Proposed BSFLA-PSO with ALL/AML Dataset
			7.5.4.4 Performance of Proposed BSFLA-PSO with ADCA Lung Dataset
			7.5.4.5 Performance of Proposed BSFLA-PSO with CNS Dataset
	7.6 Conclusion
	References
Chapter 8: Extraction of Medical Entities Using a Matrix-Based Pattern-Matching Method
	8.1 Introduction
	8.2 Background
	8.3 Methodology
		8.3.1 Dataset
		8.3.2 Proposed Method
			8.3.2.1 Text Pre-Processing
			8.3.2.2 Trained Matrix Formation
			8.3.2.3 Test Matrix Formation
			8.3.2.4 Pattern Matching
			8.3.2.5 Pruning Non-Medical Concepts
	8.4 System Evaluation
	8.5 Results and Discussion
	8.6 Conclusions and Future Work
	Acknowledgments
	References
Chapter 9: Supporting Environmental Decision MakingApplication of Machine Learning Techniques to Australia’s Emissions
	9.1 Introduction
	9.2 Data and Methodology
		9.2.1 Data
		9.2.2 Methodology
			9.2.2.1 Decision Trees
			9.2.2.2 Random Forests
			9.2.2.3 Extreme Gradient Boosting
			9.2.2.4 Support Vector Regression
		9.2.3 Data Division and the Experimental Environment
		9.2.4 Optimization of Hyperparameters
			9.2.4.1 Parameter Tuning for the DT, RF, and XGBoost Algorithms
			9.2.4.2 Parameter Tuning for the SVR Algorithm
		9.2.5 Performance Metrics
	9.3 Results and Discussion
		9.3.1 Development and Validation of the DT Model
		9.3.2 Development and Validation of the RF Model
		9.3.3 Development and Validation of the XGBoost Model
		9.3.4 Development and Validation of the SVR Model
		9.3.5 Performance Evaluation of Model
	9.4 Concluding Remarks
	References
Chapter 10: Prediction Analysis of Exchange Rate Forecasting Using Deep Learning-Based Neural Network Models
	10.1 Introduction
	10.2 Methodology
		10.2.1 Performance Measure
		10.2.2 Data Preparation
	10.3 Results and Simulations
		10.3.1 For Sliding Window Size 7
		10.3.2 For Sliding Window Size 10
		10.3.3 For Sliding Window Size 13
	10.4 Conclusion
	References
Chapter 11: Optimal Selection of Features Using Teaching-Learning-Based Optimization Algorithm for Classification
	11.1 Introduction
	11.2 Related Work
	11.3 Basic Technology
	11.4 Proposed Model
	11.5 Result Analysis
	11.6 Conclusion
	References
Chapter 12: An Enhanced Image Dehazing Procedure Using CLAHE and a Guided Filter
	12.1 Introduction
	12.2 Literature Survey
	12.3 Background Study
		12.3.1 White Balance (WB)
		12.3.2 CLAHE
		12.3.3 GF
	12.4 Proposed Methodology
	12.5 Dataset Collection and Analysis
	12.6 Image Quality Assessment Criteria
		12.6.1 Peak Signal-to-Noise Ratio and Mean Squared Error
		12.6.2 Entropy
		12.6.3 Structural Similarity Index
		12.6.4 Contrast Gain
	12.7 Experimental Results and Discussion
	12.8 Conclusion and Future Scope
	References
Index