Machine Learning with SAS® Viya®
Machine Learning with SAS® Viya® is a foundational resource for practitioners to understand the components and operation of the framework SAS Institute has adopted to deliver statistical computing, data science and numerical algorithms on cloud-based and virtual platforms. The implementation preserves and extends SAS’s 40+ year history of delivering statistical and numerical algorithms and includes up-to-date implementations of machine learning approaches – all in the virtual cloud environment. The most recent interface is both intuitive and visually appealing. The various approaches have thorough, understandable tutorials and background theory pieces. This means that an analyst can get up to speed quickly and can look back and dig down into analytical detail as required as the solution unfolds.
What is really nice about the book is that each technique has a detailed explanation of how it works and simple graphic examples are provided to demonstrate how to accomplish whatever your goal is. Overall, this demonstrates that just as these techniques have come a long way so too has the SAS environment and methodology -- producing a solution platform that incorporates best practices and is both powerful and easy to use.
Barry deVille
Data Science, IT Professional
With all the excitement about machine learning, do you feel left off the bandwagon? Are you the majority of SAS programmers who think machine learning is too complex and not worth your time? If so, I invite you to consider the ‘Machine Learning with SAS Viya’ book. For an advanced tool such as SAS Viya, you need a companion book with real-world applications to guide you through the new frontier. This book helps you to maximize your utilization of SAS Viya. Methods in this unique book show how to remove noise and redundancy so that you can focus on the clear underlying message.
The cornerstones of data, discovery and deployment are expanded to describe machine learning systems in basic non-technical terms. This book shows how to process structured and unstructured data to group, sort and transform data into more useful formats. With this method, effective decision trees, rankings and estimates are possible through visual tools. New and innovative visualization and advanced statistical modeling methods enable better understanding of the data so that pictures are worth a million words. More effective planning, mining, growing, and preventing fraud are possible. The SAS Viya tool is well optimized to handle large data volume, speed and memory, through an intuitive GUI interface with metadata information such as descriptive stats, outliers and accuracy. Both types of data can be processed to estimate p-value for scale data and absolute answers for categorical data. As with any good data science modeling, all components are applied - input, process, output, calibrate & update and repeat so the system 'learns' with each practice just like humans do. An important point to mention is that missing data can impact the machine learning process so missing data should always be quantified.
Sunil Gupta
Founder of SASSavvy.com