Название: The Handbook of Data Science and AI: Generate Value from Data with Machine Learning and Data AnalyticsАвтор: Stefan Papp, Wolfgang Weidinger, Katherine Munro, Bernhard Ortner, Annalisa Cadonna, Georg Langs, Roxane Licandro, Mario Meir-Huber
Издательство: Hanser Publications
Год: 2022
Страниц: 606
Язык: английский
Формат: epub (true)
Размер: 45.1 MB
Data Science, Big Data, and Artificial Intelligence (AI) are currently some of the most talked-about concepts in industry, government, and society, and yet also the most misunderstood. This book will clarify these concepts and provide you with practical knowledge to apply them. Featuring:
- A comprehensive overview of the various fields of application of Data Science
- Case studies from practice to make the described concepts tangible
- Practical examples to help you carry out simple data analysis projects
The book approaches the topic of Data Science from several sides. Crucially, it will show you how to build data platforms and apply Data Science tools and methods. Along the way, it will help you understand - and explain to various stakeholders - how to generate value from these techniques, such as applying Data Science to help organizations make faster decisions, reduce costs, and open up new markets. Furthermore, it will bring fundamental concepts related to Data Science to life, including statistics, mathematics, and legal considerations. Finally, the book outlines practical case studies that illustrate how knowledge generated from data is changing various industries over the long term.
There had been a dispute over whether R or Python was the primary language for data scientists. While R is popular within special user groups, especially academia, the industry seems to favor Python. As a result, Python may have become the lingua franca in this domain. Moreover, Python engineers have tons of analytical frameworks available to them, such as Keras, PyTorch, Scifi, NumPy, and many more.
Python is maintained by the Python Software Foundation and standardized by the various Enhancement Proposals (PEP). Pythonistas claim that Python is perfect for all kinds of data applications and that data engineers and data scientists should only stick to one standard. They also emphasize that Python is more readable than other languages; it is more succinct. In some cases, it takes almost just half of the code to express a routine in Python as compared to Java. Furthermore, unlike other languages, there is no compilation process needed to apply a change and execute an application again in Python.
Various engineers, however, claim that Python code is harder to debug. Especially without compiling, programmers might find problematic code in Python later than in other languages. In addition, an advocate of different languages may consider that it is harder to maintain large software projects with Python as programming languages such as Java enforce a clear structure. Or in other words: “Python allows you to create a mess if you want, while the Java compiler forces you to correct every syntax error.” Another argument is performance. The default interpreter CPython is often slower than other programming languages, and in addition, Python has language-specific bottlenecks such as the global interpreter lock (GIL) for multithreading. It is a mutex (or a lock) that allows only one thread to hold the control of the Python interpreter and which slows down multi-threaded applications.
The main reason why Python is so prevalent in the Data Science community is, perhaps, that it is easy to populate data structures for analytics. For example, in the Figure below, we just need two lines to load a custom data structure in a pandas data frame, which is a versatile data structure for all kinds of algorithms to analyze data.
Contains these current issues:
- Mathematics basics: Mathematics for Machine Learning to help you understand and utilize various ML algorithms.
- Machine Learning: From statistical to neural and from Transformers and GPT-3 to AutoML, we introduce common frameworks for applying ML in practice
- Natural Language Processing: Tools and techniques for gaining insights from text data and developing language technologies
- Computer vision: How can we gain insights from images and videos with Data Science?
- Modeling and Simulation: Model the behavior of complex systems, such as the spread of COVID-19, and do a What-If analysis covering different scenarios.
- ML and AI in production: How to turn experimentation into a working Data Science product?
- Presenting your results: Essential presentation techniques for data scientists
Contents:
Скачать The Handbook of Data Science and AI: Generate Value from Data with Machine Learning and Data Analytics