Analysis and visualisation of public procurements
Analysis of public procurement data in OCDS format.
COMMON ANOMALY RANK
METHODOLOGY
Analysis of public procurement data could be done by several approaches:
supervised, unsupervised and statistical analysis. We also developed
support for StreamStory tool, which performs data analysis of temporal
data (large multivariate event sequences).
Unsupervised analysis is based on k-Means method. Supervised analysis is
based on a decision tree analysis, and is used to get additional
insights into the public procurement decision-making process.
Statistical analysis is done in order to pursue a more intuitive and
defined-in-advance goals. StreamStory tool is used to uncover, visualize
and explain the inner structure within the data.