Building Intelligent Systems Using Machine Learning and Deep Learning: Security, Applications and Its Challenges

✍ Scribed by Abhaya Kumar, Sahoo Chittaranjan, Pradhan Bhabani, Shankar Prasad, Mishra Brojo, Kishore Mishra

• Publisher: Nova Science Publishers
• Year: 2024
• Tongue: English
• Leaves: 238
• Category: Library

✦ Synopsis

The primary objective of this book is to provide insight into the design and development of the intelligent system. The proposed book volume mainly focuses on a machine learning and deep learning-based intelligent system that would bring out the latest trends in the field of tourism, healthcare, agriculture, etc. This book provides security solutions for the intelligent system in different applications. The technological gaps between the traditional system and intelligent system are mentioned in the book, which will help in better understanding for the implementation of the intelligent system using machine learning (ML) and deep learning (DL) approaches. Although ML and DL have made great achievements in intelligent systems, there are still substantial open challenges that have not been fully studied. The main open challenges of using ML and DL in intelligent systems are: (i) Better performance of the system (ii) Time complexity of the jobs running inside an intelligent system (iii) Managing overload tasks (iv) Providing security towards the system. This book will definitely help academicians, researchers and industry people towards the security, design and development of the intelligent system.

✦ Table of Contents

Contents
Preface
Chapter 1
Intelligent Systems for Future Applications Using Machine Learning
Abstract
Introduction
Trends in Intelligent System
Key Responsabilités in Data Science towards Intelligent System
Literature Review
Applications of Machine Learning and Deep Learning Methods Used in Data Science and Intelligent System
Regression in Healthcare
Future Challenges
Classification in Agriculture
Future Challenges
Clustering for Speech Recognition
Future Challenges
Reinforcement in Self Driving Cars
Future Challenges
CNN in Facebook
Future Challenges
RNN for Face Detection
Future Challenges
Future Study and Challenges
Conclusion
References
Chapter 2
Fundamental Models in Intelligent Systems Using Machine Learning and Deep Learning
Abstract
Introduction
Related Work
Artificial Intelligence and Intelligent Systems
Machine Learning
Deep Learning
1. Linear Regression
2. Logistic Regression
3. Decision Tree
3.1. The Process of Building a Decision Tree Model Involves the Following Steps
4. Random Forest
5. Support Vector Machines (SVM)
6. Artificial Neural Networks
7. Convolutional Neural Networks
8. Recurrent Neural Networks (RNN)
9. Generative Adversarial Networks
10. Deep Belief Network
Conclusion
References
Chapter 3
A Comparative Analysis of Machine Learning Algorithms on Intrusion Detection Systems
Abstract
Introduction
Literature Review on Machine Learning in Intrusion Detection System
Proposed System
Proposed System Architecture
Packet Generation
Feature Extraction
Classification - Training and Testing
Performance Evaluation
Results and Discussion
Conclusion
References
Chapter 4
A Novel Approach for Requirement-Based Test Case Prioritization Using Machine Learning Techniques
Abstract
Introduction
Background Details
Software Testing
Regression Testing
Test Case Prioritization
Requirement Based Prioritization
Machine Learning Approaches
Classification
2.5 Cross Validation
Related Work
Discussion
Proposed Architecture
Datasets
Feature Preprocessing
Data Preparation
Data Modeling with Machine Learning Classifier
Performance Evaluation Measures
Research Questions
Experimental Result
Comparison Between Two Data Sets with or without Weight Factor
Comparison with Existing ML Techniques
Results for RQ2
Conclusion
Future Scope
References
Chapter 5
The Detection and Prevention of Phishing Threats in OSN Using Machine Learning Techniques
Abstract
Introduction
Preliminaries
OSN Threats
Classical Threats
Modern Threats
Social Threats
Combination Threats
Phishing Attack
Literature Reviews
Discussion
Proposed Methodology
Dataset
Feature Extraction and Scaling
Feature Selection
Model Building and Training
Evaluation
Experimental Results
Information Gain
Chi-Square Test
Anova Test
Solutions to Phishing Attack
Recommendation
Conclusion
Future Aspects
References
Chapter 6
A Novel Approach to Detecting Apple Disease Using CNN
Abstract
Introduction
Deep Learning
Convolutional Neural Network (CNN) Model
Optimizers
Activation Functions
Transfer Learning
1. ResNet 50
2. DenseNet 121
Literature Review
Proposed Model
Observations
Comparative Analysis
Conclusion
References
Chapter 7
A Novel Sigmoid Butterfly Optimization Deep Learning Model for Big Data Classification
Abstract
Introduction
Literature Survey
The Proposed SBOA-OGRU Technique
Apache Spark Tool
Algorithmic Design of SBOA-FS Technique
Data Classification Using OGRU Model
Experimental Validation
Epsilon Dataset
ECBDL14-ROS Dataset
Conclusion
References
Chapter 8
An Analysis of Optical Character Recognition-Based Machine Translation for Low Resource Languages
Abstract
Introduction
Applications
Review of the Existing Models
Machine Translation System Design for Low Resource Languages
Data Set Collection
Text Extraction
Pre-Processing
Encoder
Encoder Vector
Decoder
Post-processing
Output
Development of Machine Translation System for Low
Resource Languages
Dataset Preparation
Data Pre-Processing
Encode Data
Training
Encoding Phase
Decoding Phase
Sentence Prediction
Post-Processing
Rule-Based Translation
Post-Processing
Experimental Results and Analysis
Conclusion
References
Chapter 9
Generative AI for Bio-Signal Analysis
and Augmentation
Abstract
Introduction
Bio-Signal Recording Modalities
Electroencephalography
Generative AI Models
Background
Generative Adversarial Networks
Types of GANs
Deep Convolutional GAN (DCGAN)
Conditional GAN (cGAN)
Information Maximizing GAN (InfoGAN)
Auxiliary Classifier GAN (ACGAN)
Recurrent Conditional GAN (RCGAN)
WaveGAN
Evaluation of Existing GAN Networks
Methodology
Data Collection
Data Pre-Processing
GAN Models
Time-Series GANs
Application of Time-Series GAN
Model Evaluation
Challenges and Limitations
Image GANs
Application of Image GAN
Model Evaluation
Challenges and Limitations
Future of GANs
Conclusion
Disclaimer
References
Chapter 10
Deep Learning for the Closed Loop Diabetes Management System
Abstract
Introduction
Overview
Closed Loop Diabetes Management System
Role of Deep Learning in Closed Loop Diabetes Management System
Advantage of Deep Learning over Other Machine Learning Methods
Background
Diabetes Management Technology
Use of Control Systems Algorithms in Closed Loop Diabetes Management System
Machine Learning in Closed Loop Diabetes Management System
Deep Learning in Closed Loop Diabetes Management System
Methodology
Diagnosis of Diabetes
Technology in Glucose Management
Data Collection
Preprocessing
Deep Learning Architecture Selection
Model Training and Evaluation
Model Deployment
Insulin Dose Forecasting
Role of Physical Activity
Diagnosis Complications
Discussions
How Deep Learning overcome existing Techniques Technology in Glucose Management
Challenges and Limitations
Opportunities and Future Technology
Conclusion
Disclaimer
References
Chapter 11
Digital Image Spatial Feature Learning and Mapping Using Geospatial Artificial Intelligence: A Case Study
Abstract
Introduction
Methodology
Background Information and Basic Observations
Procedure of Data Map
Geoai Approach for Spatial Feature Learning and Mapping
Mapping Process Using Georeferencing
Creation of Geoai Database for Spatial Features
Learning and Mapping of Raster Data Map
Result and Discussion
Occurrence of the Final Report
Conclusion
References
About the Editors
Index
Blank Page
Blank Page