Machine Learning and IoT for Intelligent Systems and Smart Applications

Madhumathy P, M Vinoth Kumar, R. Umamaheswari, Madhumathy P, M Vinoth Kumar, R. Umamaheswari

• Editorial: CRC Press
• Año: 2021
• ISBN:
  9781000484984
• Edición: 1
• Categoría: Informatique
• Categoría: Extraction de données

The fusion of AI and IoT enables the systems to be predictive, prescriptive, and autonomous, and this convergence has evolved the nature of emerging applications from being assisted to augmented, and ultimately to autonomous intelligence. This book discusses algorithmic applications in the field of machine learning and IoT with pertinent applications. It further discusses challenges and future directions in the machine learning area and develops understanding of its role in technology, in terms of IoT security issues. Pertinent applications described include speech recognition, medical diagnosis, optimizations, predictions, and security aspects.

Features:

• 
  
  

• Focuses on algorithmic and practical parts of the artificial intelligence approaches in IoT applications.

• 
  
  

• Discusses supervised and unsupervised machine learning for IoT data and devices.

• 
  
  

• Presents an overview of the different algorithms related to Machine learning and IoT.

• 
  
  

• Covers practical case studies on industrial and smart home automation.

• 
  
  

• Includes implementation of AI from case studies in personal and industrial IoT.

This book aims at Researchers and Graduate students in Computer Engineering, Networking Communications, Information Science Engineering, and Electrical Engineering.

1 A Study on Feature Extraction and Classification Techniques for Melanoma Detection

S. Poovizhi¹, Dr. T. R. Ganesh Babu², Dr. R. Praveena³, and Dr. J. Kirubakaran⁴

¹Research Scholar, Department of Electronics and Communication Engineering, Anna University

²Professor, Department of Electronics and Communication Engineering, Muthayammal Engineering College

³Associate Professor, Department of Electronics and Communication Engineering, Muthayammal Engineering College

⁴Associate Professor, Department of Electronics and Communication Engineering, Muthayammal Engineering College

1. Contents
2. 1.1 Introduction
3. 1.2 Feature Extraction
4. 1.2.1 Fourier Transform (FT)
5. 1.2.2 Short Time Fourier Transform (STFT)
6. 1.2.3 Wavelet Transform
7. 1.2.3.1 Discrete Wavelet Transform
8. 1.2.3.2 Discrete Curvelet Transform
9. 1.2.3.3 Discrete Contourlet Transform
10. 1.2.3.4 Discrete Shearlet Transform
11. 1.2.3.5 Bendlet Transform
12. 1.3 Classification
13. 1.3.1 Logistic Regression
14. 1.3.2 K-Nearest Neighbor
15. 1.3.3 Decision Trees
16. 1.3.4 Support Vector Machine
17. 1.4 Skin Cancer Diagnostic System for Melanoma Detection
18. 1.5 Conclusion
19. References

DOI: 10.1201/9781003194415-1

1.1 Introduction

The incidence of malignant melanoma is in the majority cases fatal and increasing worldwide. According to the 2020 Melanoma Skin Cancer Report of the Global Cancer Observatory, there were 287,723 cases of melanoma and 1,042,056 of non-melanoma cancers recorded globally with a greater number of cases in Australia and the United States of America than anywhere else in the world [1]. Gender-wise, men are 10% more likely to develop melanoma skin cancer than women and 4% more likely to die from melanoma than women. (Figure 1.1) gives a snapshot of growth of skin cancer from current to future projected levelsand (Table 1.1) gives projected levels of new cases of skin cancer.

Table 1.1
Global Cancer Observatory Report 2020

Estimated New Cases in 2025	340,271
Estimated New Cases in 2040	466,914

Figure 1.1 Incidence and Mortality Rate of Melanoma Cancer.

The two major types of skin cancers are Melanoma and Non-Melanoma. Melanoma arises from malignant melanocytic cells of the epidermis and this cell produce melanin pigment which decides the color of the skin. The malignant melanocytic cells grow abnormally and invade other skin cells forming a big mass of cells called the tumor. Early detection and treatment profoundly lead to prognosis of the disease. With many skin imaging techniques developed in recent years for assisting dermatologists to detect the melanoma, some of the techniques are (i) Total Body Photograph [2] (ii) Ultrasonography [3] (iii) Epiluminescence Microscopy [4] (iv) Cross Polarization Epiluminescence...