Название: Machine Learning: Concepts, Tools And Data VisualizationАвтор: Minsoo Kang, Eunsoo Choi
Издательство: World Scientific Publishing
Год: 2021
Страниц: 296
Язык: английский
Формат: pdf (true)
Размер: 118.4 MB
This set of lecture notes, written for those who are unfamiliar with mathematics and programming, introduces the reader to important concepts in the field of Machine Learning. It consists of three parts. The first is an overview of the history of Artificial Intelligence (AI), Machine Learning (ML), and Data Science (DS), and also includes case studies of well-known AI systems. The second is a step-by-step introduction to Azure Machine Learning, with examples provided. The third is an explanation of the techniques and methods used in data visualization with R, which can be used to communicate the results collected by the AI systems when they are analyzed statistically. Practice questions are provided throughout the book.
This book, firstly, was written for those who do not know mathematics and the theory of control, secondly, for those who tried the open-source approach but had difficulties.
People who don’t know mathematics or control theory will be told what AI is and how data can be used in the application of AI. For best results, they should follow the instructions in the book. Those who use open-source approaches can read the text carefully and decide whether to choose the open-source approach or not. There are pros and cons to choosing an open-source approach. Because it is open-source, the approach will cost nothing, but there will be limitations in applying the required libraries or algorithms. The time spent finding alternative ways to perform tasks could be better invested in actual research. The difference between closed-source and open-source approaches is similar to the difference between Windows/iOS users and Linux users. Windows/iOS users pay for their OS but have a greater variety of programs and access to support. Linux users, on the other hand, do not pay for their OS, but their OS is less convenient and more difficult to use. I hope you understand my analogy explaining the difference between open-source and closed-source approaches. For this book, I have used Microsoft’s Azure Machine Learning program.
This book consists of three parts. Part 1 briefly introduces the history of Artificial Intelligence, Machine Learning, and Deep Learning, as well as algorithms. It also explains how the data will be utilized i.e. in data science approaches such as data mining and big data processing. Data mining and big data processing are ways of discovering patterns in large data sets involving methods at the intersection of Machine Learning, statistics, and database systems.
Part 2 introduces Azure Machine Learning, how to use it, and some examples as needed. Azure Machine Learning makes it easy to use algorithms used in supervised and unsupervised learning by selecting menus. So, Azure Machine Learning is explained and each step is illustrated so that we can study the example program from the beginning to the end. We also described examples of using actual public and medical data. If this book is read in its entirety, even beginners who have just started machine learning can use Azure Machine Learning programs.
Part 3 introduced visualization and Power BI, which is a visualization tool. Data visualization is an essential technique in all areas of data handling. Visualization will help you lay the foundation of data science. Power BI was developed by Microsoft, and so can be integrated with Microsoft Azure Machine Learning. Also, Power BI’s R Script feature enables complex computations for high-performance visualization.
If you want to apply Artificial Intelligence rather than follow the theoretical approach, you will have to learn a program, not necessarily Azure Machine Learning Studio. Local and overseas laboratories have developed many Artificial Intelligence programs, but the Azure ML Studio program, chosen by the author, offers a wide variety of functions, which will be required in this book. However, you need to know that licensing costs are incurred. Here’s a tip for beginners who are starting Azure Machine Learning: we recommend that you take advantage of the 8-hour trial version. Readers should expect to be able to understand AI and visualize the results of using AI on their data by the end of the book.
Скачать Machine Learning: Concepts, Tools And Data Visualization