Intel® DAAL accelerates the entire data analysis workflow, from data acquisition, to data transformation, to data analysis (mining), training, and prediction.
Data acquisition includes the streaming of out-of-memory dataset, which may include data decompression. Hence the DAAL covers important data sources such as CSV files and SQL queries as well as general compression and decompression method such as LZO, RLE, etc.
DAAL optimizes data management of both homogeneous and non-homogeneous data (numerical and categorical).
Data transformation includes variety of data aggregation and dimension reduction techniques such as Principal Component Analysis, Singular Value Decomposition, etc.
Data analysis includes algorithms for Summary Statistics and Clustering.
The last but not least, DAAL optimizes several important Machine Learning algorithms, both training and prediction stages, which include regression (Linear Regression), classification (Boosting, Support Vector Machines), and Association Rules.