Abstract:
|
A Trinary or Trinomial-based orientation to Big Data analytics has the advantage of identifying relevant regions within large datasets for further exploration. This approach also formulates subsets or clusters which can be linked to multiple parameters. The Trinary (-/0/+) regions inherent in nominal or ordinal based distributions essentially characterize input records feasible for multi-parameter linking. The authors have demonstrated applications to air quality monitoring and Geographic Information Science (GIS) based depiction of lung-cancer mortality rates. This paper deals with applications to international indicators such as ethnic and linguistic fractionalization, GINI index, GDP per capita USD, and Happiness Index derived from the GALLUP world poll. Preliminary analytical findings from a multivariate model interfaced with derived clusters are presented.
|