Intel® Data Analytics Acceleration Library provides a collection of popular algorithms for performing data analysis and machine learning in variety of usage scenarios, including offline, online, and distributed computing usages.
For example, if you work with high dimensional data you might find useful the algorithms intended for dimension reduction, such as the Principal Component Analysis or Singular Value Decomposition.
For data structure discovery you might want to use summary statistics algorithms and/or clustering techniques available in the library.
For modeling you might want to use some available methods for regression and classification.
But the Intel DAAL is more than Data Mining/Machine Learning library. It provides you the entire framework for data management: data acquisition from a data source, compression and decompression, data streaming into memory, effective in-memory data management, serialization and deserialization, etc.
Download the Intel DAAL 2016 Beta and try it! Comprehensive API Reference and Programming Guide will help you to learn about library algorithms and usage scenarios.