Page 33 - Invited Paper Session (IPS) - Volume 1

IPS35 Dinov I.D. et al.
applications using neurodegenerative aging disorders, paediatric pathological
brain development, and exploratory census-like population neuroscience.
Figure 1 shows a schematic that illustrates the relation between census-
like population-based view of natural processes (left), their Big Data proxy
representation (middle), and classical (small) sampling based process
description. By examining many dozens of complex biomedical and health
case-studies we identified the common characteristics of Big Data (Dinov
2018), Table 1.

Figure 1: Schematic of the relation between native processes (left), their Big
Data representations (middle), and traditional sampling based process
characterization. Note that (1) the ideal population view of the process is often
unobservable and intractable, the Big Data proxy of the process often requires
substantial data management, harmonization, aggregation, preprocessing
and wrangling before it can be analysed, and (3) the sample data may facilitate
rapid and effective data analytics, but may also represent a limited view of the
entire process.

22 | I S I W S C 2 0 1 9