Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Big data --- Data management

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Enhance your SAS data-wrangling skills with high-precision and parallel data manipulation using the DS2 programming language. Now in its second edition, this book addresses the DS2 programming language from SAS, which combines the precise procedural power and control of the Base SAS DATA step language with the simplicity and flexibility of SQL. DS2 provides simple, safe syntax for performing complex data transformations in parallel and enables manipulation of native database data types at full precision. It also covers PROC FEDSQL, a modernized SQL language that blends perfectly with DS2. You will learn to harness the power of parallel processing to speed up CPU-intensive computing processes in Base SAS and how to achieve even more speed by processing DS2 programs on massively parallel database systems. Techniques for leveraging internet APIs to acquire data, avoiding large data movements when working with data from disparate sources, and leveraging DS2's new data types for full-precision numeric calculations are presented, with examples of why these techniques are essential for the modern data wrangler. Here's what's new in this edition: how to significantly improve performance by using the new SAS Viya architecture with its SAS Cloud Analytic Services (CAS) how to declare private variables and methods in a package the new PROC DSTODS2 the PCRXFIND and PCRXREPLACE packages While working though the code samples provided with this book, you will build a library of custom, reusable, and easily shareable DS2 program modules, execute parallelized DATA step programs to speed up a CPU-intensive process, and conduct advanced data transformations using hash objects and matrix math operations. This book is part of the SAS Press Series.

Data mining. --- Big data.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Over the past few years big data has emerged a new major pluri-disciplinary research area whose objective is to deal with massive datasets and that are too large to be manipulated by conventional modelling and current computational approaches. As the datasets often available in many disciplines and application areas exponentially grow thanks to the increasing availability of novel sensors and devices, there is nowadays an urgent need to develop novel methods and computational approaches to deal with the high volumes and variety of the large data sources generated in many fields. The range of research challenges still opened are extremely large, from the integration, storage, manipulation, data mining and visualization techniques to the development of computing architectures, cloud and distributed computing platforms to scalable storage systems. Clearly, big data challenges should require multidisciplinary approaches, when different points of view, ideas and experiences should be shared, this favoring exchanges and cross-fertilization on recent trends as well as research directions to consider.

Big data --- Data mining

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Data mining

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Data mining