Highlighted Selections from:

Data Mining and Official Statistics: the Past, the Present and the Future

DOI: 10.1089/big.2013.0038

Hassani, Hossein, Gilbert Saporta, and Emmanuel Sirimal Silva. “Data Mining and Official Statistics: the Past, the Present and the Future.” Big Data (2014): 1-10. Web.

p.1: The minimal applications of data mining for official statistics are not entirely surprising for the following reasons: first, National Statistical Institutes (NSIs) are tasked with data collection, while the common practice has been to outsource the analysis; second, the objective of official statisticians is to answer precise questions and make forecasts as opposed to finding unexpected patterns or models. -- Highlighted mar 12, 2014

p.2: However, it should be noted that data mining is not concerned with efficient methods for collecting data such as surveys and experimental designs. Furthermore, models do not come from a theory, but from data exploration. As such, data mining is not concerned with estimation and tests or prespecified models but with discovering models through an algorithmic search process exploring linear and nonlinear models, explicit or not. -- Highlighted mar 12, 2014

p.3: the lucrative application of data mining should be to use techniques that integrate and appreciate privacy concerns. "Symbolic data analysis" was introduced in order to overcome the challenges imposed by aggregated data. -- Highlighted mar 12, 2014

p.4: data mining now uses security control mechanisms known as query restriction or noise addition, to prevent the revelation of confidential individual information while safeguarding the data quality. -- Highlighted mar 12, 2014

p.6: there is substantial evidence that government statisticians are trapped in traditions that limit their exposure and willingness to exploit and explore lucrative and novel data mining techniques that can improve and enhance their efficiency and quality of information provided through official statistics. The answer to this challenge is the second point following from the analysis, which compels us to reiterate the call for increased engagement, cooperation, and collaboration -- Highlighted mar 12, 2014