The gCube Data Analysis Package is a a statistical library developed in Java programming language, which provides gCube with facilities for the following aims:

• Perform data processing and mining;
• Produce data in a common format;
• Provide a standalone, open source, library that can be used even by third parties
• Provide a library which aggregates state-of-the-art modules for statistical analysis and machine learning techniques;
• Build up a little "desktop laboratory" for data processing, which supports offline charts visualization for "testing before deploying.

Currently, implemented functionalities are:

Multiple data representation: A module for charts visualization is currently included, as well as classes for performing data transformation

• Data Mining;
• Standalone Analysis;
• Transects Extraction;
• Set Identity Guessin.

The statistical library functionalities are increasing more and more for including other non-deterministic operations which can be performed on timeseries and statistical data.