Once you decide to implement a data security strategy, it can be difficult to know where to start. With so many potential threats and challenges to resolve, teams often try to fix everything at once. But this boil-the-ocean approach is difficult to manage efficiently and ultimately leads to frustration, confusion, and halted progress. There's a better way to go. In this report, data science and AI leader Federico Castanedo shows you what to look for in a data security platform that will deliver the speed, scale, and agility you need to be successful in today's fast-paced, distributed data ecosystems. Unlike other resources that focus solely on data security concepts, this guide provides a road map for putting those concepts into practice. This report reveals: The most common data security use cases and their potential challenges What to look for in a data security solution that's built for speed and scale Why increasingly decentralized data architectures require centralized, dynamic data security mechanisms How to implement the steps required to put common use cases into production Methods for assessing risks—and controls necessary to mitigate those risks How to facilitate cross-functional collaboration to put data security into practice in a scalable, efficient way You'll examine the most common data security use cases that global enterprises across every industry aim to achieve, including the specific steps needed for implementation as well as the potential obstacles these use cases present. Federico Castanedo is a data science and AI leader with extensive experience in academia, industry, and startups. Having held leadership positions at DataRobot and Vodafone, he has a successful track record of leading high-performing data science teams and developing data science and AI products with business impact.

The amount of data generated from various processes and platforms has increased exponentially in the past decade, and the challenges of filtering useful data out of streams of raw data has become even greater. Meanwhile, the essence of making useful insights from that data has become even more important. In this incisive report, Federico Castanedo examines the challenges companies face when acting on data at rest as well as the benefits you unlock when acting on data as it's generated. Data engineers, enterprise architects, CTOs, and CIOs will explore the tools, processes, and mindset your company needs to process streaming data in real time. Learn how to make quick data-driven decisions to gain an edge on competitors. This report helps you: Explore gaps in today's real-time data architectures, including the limitations of real-time analytics to act on data immediately Examine use cases that can't be served efficiently with real-time analytics Understand how stream processing engines work with real-time data Learn how distributed data processing architectures, stream processing, streaming analytics, and event-based architectures relate to real-time data Understand how to transition from traditional batch processing environments to stream processing Federico Castanedo is an academic director and adjunct professor at IE University in Spain. A data science and AI leader, he has extensive experience in academia, industry, and startups.

Organizations are rapidly consuming more data than ever before, and to drive their competitive advantage, they’re demanding interactive visualizations and interactive analyses of that data be embedded in their applications and business processes. This will enable them to make faster and more effective decisions based on data, not guesses. This practical book examines the considerations that software developers, product managers, and vendors need to take into account when making visualization and analytics a seamlessly integrated part of the applications they deliver, as well as the impact of migrating their applications to modern data platforms. Authors Federico Castanedo (Vodafone Group) and Andy Oram (O’Reilly Media) explore the basic requirements for embedding domain expertise with fast, powerful, and interactive visual analytics that will delight and inform customers more than spreadsheets and custom-generated charts. Particular focus is placed on the characteristics of effective visual analytics for big and fast data. Learn the impact of trends driving embedded analytics Review examples of big data applications and their analytics requirements in retail, direct service, cybersecurity, the Internet of Things, and logistics Explore requirements for embedding visual analytics in modern data environments, including collection, storage, retrieval, data models, speed, microservices, parallelism, and interactivity Take a deep dive into the characteristics of effective visual analytics and criteria for evaluating modern embedded analytics tools Use a self-assessment rating chart to determine the value of your organization’s BI in the modern data setting

One viable option for organizations looking to harness massive amounts of data is the data lake, a single repository for storing all the raw data, both structured and unstructured, that floods into the company. But that isn’t the end of the story. The key to making a data lake work is data governance, using metadata to provide valuable context through tagging and cataloging. This practical report examines why metadata is essential for managing, migrating, accessing, and deploying any big data solution. Authors Federico Castanedo and Scott Gidley dive into the specifics of analyzing metadata for keeping track of your data—where it comes from, where it’s located, and how it’s being used—so you can provide safeguards and reduce risk. In the process, you’ll learn about methods for automating metadata capture. This report also explains the main features of a data lake architecture, and discusses the pros and cons of several data lake management solutions that support metadata. These solutions include: Traditional data integration/management vendors such as the IBM Research Accelerated Discovery Lab Tooling from open source projects, including Teradata Kylo and Informatica Startups such as Trifacta and Zaloni that provide best of breed technology

One area where data analytics can have profound effect is your company’s procurement process. Some organizations spend more than two thirds of their revenue buying goods and services, making procurement—out of all business activities—a key element in achieving cost reduction. This report examines how your company can significantly improve procurement analytics to solve business questions quickly and effectively. Author Federico Castanedo, Chief Data Scientist at WiseAthena.com, explains how a probabilistic, bottom-up approach can significantly increase the quality, speed, and scalability of your data preparation operations—whether you’re integrating datasets or cleaning and classifying them. You’ll learn how new solutions leverage automation and machine learning, including the Tamr platform, and help you take advantage of several data-driven actions for procurement—including compliance, price arbitrage, and spend recovery.

Preparing and cleaning data is notoriously expensive, prone to error, and time consuming: the process accounts for roughly 80% of the total time spent on analysis. As this O’Reilly report points out, enterprises have already invested billions of dollars in big data analytics, so there’s great incentive to modernize methods for cleaning, combining, and transforming data. Author Federico Castanedo, Chief Data Scientist at WiseAthena.com, details best practices for reducing the time it takes to convert raw data into actionable insights. With these tools and techniques in mind, your organization will be well positioned to translate big data into big decisions. Explore the problems organizations face today with traditional prep and integration Define the business questions you want to address before selecting, prepping, and analyzing data Learn new methods for preparing raw data, including date-time and string data Understand how some cleaning actions (like replacing missing values) affect your analysis Examine data curation products: modern approaches that scale Consider your business audience when choosing ways to deliver your analysis