Название: Confident Data Skills: How to Work with Data and Futureproof Your Career, 2nd Edition
Автор: Kirill Eremenko
Издательство: Kogan Page
Серия: Confident Series
Год: 2020
Страниц: 321
Язык: английский
Формат: pdf (true), epub
Размер: 10.1 MB

Data has dramatically changed how our world works. Understanding and using data is now one of the most transferable and desirable skills. Whether you're an entrepreneur wanting to boost your business, a jobseeker looking for that employable edge, or simply hoping to make the most of your current career, Confident Data Skills is here to help.

This updated second edition takes you through the basics of dаta: from data mining and preparing and analysing your data, to visualizing and communicating your insights. It now contains exciting new content on neural networks and Deep Learning (DL). Featuring in-depth international case studies from companies including Amazon, LinkedIn and Mike's Hard Lemonade Co, as well as easy-to understand language and inspiring advice and guidance, Confident Data Skills will help you use your new-found data skills to give your career that cutting-edge boost.

By now, you’re likely to have heard the term 'Big Data'. Put very simply, Big Data is the name given to datasets with columns and rows so considerable in number that they cannot be captured and processed by conventional hardware and software within a reasonable length of time. For that reason, the term is dynamic – what might once have been considered Big Data back in 2015 will no longer be thought of as such in 2022, because by that time technology will have been developed to tackle its magnitude with ease.

Narrow AI: The term can be misleading, as 'narrow' suggests severe limitations. Don’t let it fool you – narrow AI runs the gamut of all intelligent processes, from the spam detection algorithm in your email inbox to self-driving cars that can track objects, decide when to turn and distinguish road signs. Because of the ongoing expansion of the field, it is difficult to find a stable taxonomy for AI. We shall therefore discuss the five most significant technologies in the field:

- Robotic process automation (RPA)
- Computer vision (CV)
- Natural language processing (NLP)
- Reinforcement learning (RL)
- Deep learning (DL)

The Data Science Process leads us through every stage of our project, from the moment we first consider how to approach the data, to presenting our findings in a clear and actionable way. The process has five stages, which are:

- Identify the question.
- Prepare the data.
- Analyse the data.
- Visualize the insights.
- Present the insights.

In the first edition of this book, Python and R were on a roughly equal playing field. With little differences between them I recommended that data scientists choose one or the other and specialize in that tool. This has since changed – with Python clearly taking the lead. Samuel Hinton notes that while some in his community use R for analysis, the problem lies in its breadth of use. ‘Outside of academia, Python has you covered but R not so much. Python is popular in almost every area that uses coding and data analysis’. Python has two clear advantages. The first is in relation to deep learning, which is taking the world of analytics by storm – Python is an agile, powerful creation tool in this space. While R also offers tools for deep learning, Python takes the lead because both Facebook’s PyTorch and Google’s Tensorflow open-source deep learning libraries are based on this programming language. The second advantage is that, unlike R, Python is not limited to data analytics, but instead has a much broader scope of applications: among others, software development, web development, and even to enable 3D applications.

"The most comprehensive book I have seen for those wanting to get into data science - what Harvard Business Review called 'the sexiest job of the 21st century'."--Ben Taylor, Chief AI Evangelist, DataRobot

"Kirill Eremenko's book skilfully unravels the mysteries behind all the popular analytics tools and techniques, as well as many of the algorithms that power intelligent systems. I would recommend it to anyone who wants to pursue a career in data science. "--Dan Shiebler, Senior Machine Learning Engineer, Twitter Cortex