Book Image

Managing Data Science

By : Kirill Dubovikov
Book Image

Managing Data Science

By: Kirill Dubovikov

Overview of this book

Data science and machine learning can transform any organization and unlock new opportunities. However, employing the right management strategies is crucial to guide the solution from prototype to production. Traditional approaches often fail as they don't entirely meet the conditions and requirements necessary for current data science projects. In this book, you'll explore the right approach to data science project management, along with useful tips and best practices to guide you along the way. After understanding the practical applications of data science and artificial intelligence, you'll see how to incorporate them into your solutions. Next, you will go through the data science project life cycle, explore the common pitfalls encountered at each step, and learn how to avoid them. Any data science project requires a skilled team, and this book will offer the right advice for hiring and growing a data science team for your organization. Later, you'll be shown how to efficiently manage and improve your data science projects through the use of DevOps and ModelOps. By the end of this book, you will be well versed with various data science solutions and have gained practical insights into tackling the different challenges that you'll encounter on a daily basis.
Table of Contents (18 chapters)
Free Chapter
Section 1: What is Data Science?
Section 2: Building and Sustaining a Team
Section 3: Managing Various Data Science Projects
Section 4: Creating a Development Infrastructure

Deep learning use case

To show how deep learning may work in practical settings, we will explore product matching.

Up-to-date pricing is very important for large internet retailers. In situations where your competitor lowers the price of a popular product, late reaction leads to large profit losses. If you know the correct market price distributions for your product catalog, you can always remain a step ahead of your competitors. To create such a distribution for a single product, you first need to find this product description on a competitor's site. While automated collection of product descriptions is easy, product matching is the hard part.

Once we have a large volume of unstructured text, we need to extract product attributes from it. To do this, we first need to tell whether two descriptions refer to the same product. Suppose that we have collected a large dataset of...