Hybrid Computational Intelligence: Challenges and Applications

Hybrid Computational Intelligence
Copyright
Dedication
Contents
List of contributors
Preface
1 Application and techniques of opinion mining
	1.1 Introduction
	1.2 Fundamentals of opinion mining
		1.2.1 Defining opinion
		1.2.2 Defining sentiments
		1.2.3 Task of opinion mining
		1.2.4 Representation of opinion
	1.3 Feature extraction and its impact on opinion mining
		1.3.1 Analyzing text and fact reviews
			1.3.1.1 Analysis and classification of features
			1.3.1.2 Analysis and categorization of sentiments
	1.4 Deep learning and its relation to opinion mining
		1.4.1 Opinion mining with deep learning
	1.5 Techniques of opinion mining
		1.5.1 Lexicon-based approach
		1.5.2 Naive Bayes classifier
		1.5.3 Support vector machines
		1.5.4 Decision trees
	1.6 Tools of opinion mining
		1.6.1 WEKA
			1.6.1.1 Features of WEKA
			1.6.1.2 Installation of WEKA
			1.6.1.3 WEKA application interfaces
		1.6.2 Apache OpenNLP
	1.7 Ontology-based opinion mining
	1.8 Applications of opinion mining
	1.9 Conclusion
	References
2 Influence of big data in smart tourism
	2.1 Introduction to smart tourism
		2.1.1 Introduction
		2.1.2 Smart cities are leading the way for smart tourism
	2.2 Introduction to big data
		2.2.1 What is big data?
		2.2.2 Big data systems are different
		2.2.3 Types of big data
		2.2.4 Life cycle of big data
	2.3 Applications of big data
		2.3.1 Health care
		2.3.2 Media and entertainment
		2.3.3 Internet of Things
		2.3.4 Manufacturing
		2.3.5 Government
	2.4 Use of big data in the travel sector
	2.5 Big data is transforming the travel industry
	2.6 Key findings for the travel sector
		2.6.1 Benefits of big data to tourism businesses
	2.7 Tools for big data analysis for smart tourism
		2.7.1 Open source data analytics tools
		2.7.2 Demonstration of R, Spyder, and Jupyter Notebook
			2.7.2.1 Using Spyder and R
			2.7.2.2 Using Jupyter Notebook
		2.7.3 Analysis of the tourism data set
	2.8 Applying PROPHET and ARIMA prediction models
		2.8.1 PROPHET model
		2.8.2 ARIMA model
		2.8.3 Analysis of the prediction models
	2.9 Challenges in data-intensive tourism
	2.10 Conclusion and future scope
	References
	Further reading
3 Deep learning and its applications for content-based video retrieval
	3.1 Introduction
	3.2 Video retrieval techniques
		3.2.1 Feature-based video retrieval
		3.2.2 Spatial analysis
		3.2.3 Spatiotemporal querying in video databases
	3.3 Video querying
		3.3.1 Visual query
		3.3.2 Motion query
		3.3.3 Textual query
		3.3.4 Feature clustering
			3.3.4.1 Method
	3.4 Deep learning for video analysis
		3.4.1 Convolutional neural network for content-based video retrieval
		3.4.2 Feature analysis using convolutional neural network for video data
	3.5 A multitier deep learning-based video classification using C3D
		3.5.1 Data set
		3.5.2 Methodology
		3.5.3 Experiment and results
	References
4 A computationally intelligent agent for detecting fake news using generative adversarial networks
	4.1 Fake news
		4.1.1 What is fake news?
		4.1.2 Reasons for the emergence of fake news
		4.1.3 Effects and dangers of fake news
	4.2 Deep learning
		4.2.1 What is deep learning?
		4.2.2 Introduction to neural networks and gradient descent
		4.2.3 Deep learning architectures
			4.2.3.1 Recurrent neural networks
			4.2.3.2 Long short-term memory/gated recurrent unit
			4.2.3.3 Convolutional neural network
			4.2.3.4 Deep belief networks
			4.2.3.5 Deep stacking networks
		4.2.4 Applications of deep learning
			4.2.4.1 Computer vision
			4.2.4.2 Speech recognition
			4.2.4.3 Natural language processing
	4.3 Generative adversarial networks
		4.3.1 What are generative adversarial networks?
		4.3.2 Why should we use generative adversarial networks?
		4.3.3 Applications of generative adversarial network
	4.4 A case study on generative adversarial networks
		4.4.1 Basics of generative adversarial networks
			4.4.1.1 The architecture
			4.4.1.2 The generator
			4.4.1.3 The discriminator
			4.4.1.4 The adversarial competition
			4.4.1.5 The math
		4.4.2 Using generative adversarial networks for generating synthetic images
			4.4.2.1 The experiment
			4.4.2.2 The simple architecture
			4.4.2.3 The convolutional neural network architecture
			4.4.2.4 The training
			4.4.2.5 Results
		4.4.3 Implications of results for fake news detection
		4.4.4 Problems with generative adversarial networks for text data
		4.4.5 Introduction to SeqGAN, a generative adversarial network architecture for text data
	4.5 Experiment and results
		4.5.1 Setup
		4.5.2 Process
		4.5.3 Results
	4.6 Summary
	References
	Further reading
5 Hybrid computational intelligence for healthcare and disease diagnosis
	5.1 Introduction
		5.1.1 A brief description of hybrid intelligence
			5.1.1.1 Example hybrid intelligence: hybrid neuro-expert system
			5.1.1.2 Overview of a hybrid intelligent system: neural expert system
			5.1.1.3 Evolution of computational intelligence in health care
			5.1.1.4 Healthcare safety issues
	5.2 Medical image segmentation and classification
		5.2.1 Medical image segmentation
			5.2.1.1 Thresholding
			5.2.1.2 Edge-based segmentation
			5.2.1.3 Region-based segmentation
			5.2.1.4 Pixel classification
		5.2.2 Medical image classification
			5.2.2.1 Supervised classification
			5.2.2.2 Unsupervised classification
		5.2.3 Medical image classification based on a hybrid approach
			5.2.3.1 Neuro fuzzy system
			5.2.3.2 Neural expert systems
		5.2.4 Automated fuzzy clustering for biomedical images
		5.2.5 Segmentation of magnetic resonance imaging images using neural network techniques
	5.3 Disease and diagnosis approach
		5.3.1 Hybrid computational intelligence for the prediction of breast cancer
			5.3.1.1 Breast cancer detection and artificial neural networks
			5.3.1.2 Intelligent techniques and breast cancer detection
		5.3.2 Genetic algorithms and partial swarm optimization
			5.3.2.1 Genetic algorithms
			5.3.2.2 Particle swarm optimization
	5.4 Identifying brain activity using a state classifier
		5.4.1 Modeling the data science for a diagnosis technique
	5.5 Genomics
		5.5.1 Method for gene functional enrichments
			5.5.1.1 Gene enrichment score
			5.5.1.2 Statistical estimation of gene enrichment score
			5.5.1.3 Multiple hypothesis testing
		5.5.2 Modern artificial file
	5.6 Health bioinformatics
		5.6.1 Multimodal medical image fusion
		5.6.2 Recognition critical genes in Austin syndrome
	5.7 Discussion
	5.8 Conclusion
	References
6 Application of hybrid computational intelligence in health care
	6.1 Introduction
		6.1.1 Computational intelligence
		6.1.2 Hybrid computational intelligence
		6.1.3 Health care
	6.2 Need for computational intelligence in health care
	6.3 Need for hybrid computational intelligence in health care
	6.4 Use cases for hybrid computational intelligence in health care
		6.4.1 Research and clinical decision support
		6.4.2 Medical imaging
		6.4.3 Voice to text transcription
		6.4.4 Fraud detection
			6.4.4.1 Data for healthcare fraud
			6.4.4.2 Methods for healthcare fraud detection
		6.4.5 Cyber security
			6.4.5.1 Cyber security challenges in health care
			6.4.5.2 Strategies for improving cyber security
	6.5 Conclusion
	References
7 Utility system for premature plant disease detection using machine learning
	7.1 Introduction
	7.2 Literature survey
		7.2.1 Related work
			7.2.1.1 Image processing
			7.2.1.2 Feature extraction
			7.2.1.3 Classification and clustering
		7.2.2 Current system
		7.2.3 Proposed system
	7.3 Design and implementation
	7.4 Results
	7.5 Conclusion
	References
	Further reading
8 Artificial intelligence-based computational fluid dynamics approaches
	8.1 Introduction
	8.2 AI-based computational fluid dynamic approaches
		8.2.1 The use of artificial neural networks in computational fluid dynamic expert systems
		8.2.2 Coupled artificial intelligence (via artificial neural network) and computational fluid dynamic in predicting heat ex...
		8.2.3 Fluid flow optimization using artificial intelligence-based convolutional neural networks computational fluid dynamic
		8.2.4 Genetic algorithm-based computational fluid dynamic multiobjective optimization
		8.2.5 The use of an Elman neural network as an artificial intelligence in computational fluid dynamic hull form optimization
	8.3 Conclusion
	References
	Appendix
9 Real-time video segmentation using a vague adaptive threshold
	9.1 Introduction
	9.2 Temporal video segmentation (shot boundary detection)
		9.2.1 Shot transition detection methods based on classical approaches
		9.2.2 Soft computing-based approaches for shot transition detection
		9.2.3 Methods for real-time shot boundary detection
	9.3 Basic concepts and preliminaries
		9.3.1 Fuzzy hostility map generation using spatio-temporal fuzzy hostility index
		9.3.2 Pixel intensification and fuzzy hostility map dilation
		9.3.3 Fuzzy hostility map similarity
		9.3.4 Vague set modeling for video segmentation
		9.3.5 Computation of vague adaptive threshold (VAT) in real-time
		9.3.6 Look-back technique for detection of missed video segments
	9.4 Approach for real-time video segmentation
		9.4.1 Extraction of features from buffered frames
		9.4.2 Computation of VAT for detection of shot boundaries in real-time
		9.4.3 Detection of missed video segments using the look-back technique
	9.5 Experimental results and analysis
		9.5.1 Video dataset used for experimentation
		9.5.2 Experimental results
		9.5.3 Comparison with other existing methods
	9.6 Future directions and conclusion
	References
Index