There are two main targets for future work with regard to trend analysis and pattern detection as they are used in HCPP interactive visualization tools. The first is to incorporate iterated function systems (IFS) and fractal subdivision. IFS provide a structured approach to synthesize "real-world" data for testing of forecasting tools. Fractal subdivision, on the other hand, can be used to "intelligently" fill the gaps in data sets with missing samples by using characteristics of the samples that are present.
A second target for future work is to support the estimation of the maximum Lyapunov exponent (MLE) in HCPP data visualization tools. The MLE can be used to establish how chaotic a dynamic system may be. In essence, a measure of how difficult a data set may be to forecast can be determined before blind predictions are made, regardless of whether those prediction models are widely accepted as appropriate, standardized models. A discussion of the MLE can be found here.