Abstract:
|
The question was how the explosion in the availability of health- and disease-related data from biological, biomedical, behavioral, social, environmental, and economical analyses could be addressed in view of the analysis of biomedical big data. And this was the challenge posed by complexity of data structures such as images, networks, and graphs, missing and sparse data, and complex dependence structures and interaction effects. The new method is based on changes of Cumulative Distribution Function in relation to time change in sampling patterns. Multidimensional Time Model for Probability Cumulative Function can be reduced to finite-dimensional time model, which can be characterized by Boolean algebra for operations over events and their probabilities and index set for reduction of infinite dimensional time model to finite number of dimensions of time model considering also the fractal-dimensional time arising from alike supersymmetrical properties of probability. The applicability of results are further extended to be used in innovative methodology for visualization, modeling, and analysis of biomedical big data to address the challenges posed by complex data structures such as images, networks, and graphs, missing and sparse data, and complex dependence structures and interaction effects such as various DNA analysis. The newly developed models, philosophically based on Erdos- Reney Law for the prediction are philosophically intended to reach high level of precision.
|