JMSL Numerical Library Version 3.0 for Java applications offers a broad collection of mathematical, statistical, financial and charting classes and provides a scalable framework for tailored analytical applications, enabling users in finance, business intelligence, bioinformatics and life science to create customized applications for data mining and predictive analysis of complex data. Features include neural network engines for advanced forecasting and data mining, algorithms for pre- and post-processing of data, added multiple linear regression and non-linear regression algorithms for producing predictive models, and hierarchical cluster analysis for classifying and analyzing data. The library includes neural network technology that adds to the selection of data mining, modeling and prediction technologies available across the family of products. The neural network technology mimics human problem-solving processes by applying knowledge gained from historical data to new situations, enabling businesses to build predictive models that increase in accuracy as more information is obtained. An important addition, the heat map chart type is most commonly used in life sciences to display two-dimensional arrays of color values.