Skip to main content

VVV: Discovering the possibilities of Big Data

Our world is full of Big Data, which, if correctly analyzed, have the potential to lead us to a better understanding of several social and economic phenomena.  Big Data can be described as data sets characterized by 3Vs:


Volume = impossible to process it with traditional database systems and computer environments
Velocity = analysis of streaming data – the picture is constantly changing
Variety = many different forms of data, often unstructured

Since January 2014, the UNECE Statistical Division has been coordinating a major international collaboration project on exploring the potential of Big Data as a source for official statistics. Within this framework, some experimental studies have been carried on tourism trends by using mobile positioning data, exploiting internet traffic flows for collecting statistics on the Information Society…  Here are some of the emerging areas of study. 


 


 



 Mobility patterns to touristic cities in Mexico
(map generated from geo-referenced data of 42 million single tweets)



 


The project shows for the first time, on a practical basis and a broad scale, the potential and the limits of the use of Big Data as sources for computing statistics. It is clear that in some areas, Improvements in efficiency and quality are possible by replacing current data sources with Big Data.


 



 


New products can be obtained from novel sources such as traffic detectors, mobile phones and social media data. However, some sources can be of low quality, and require some serious pre-processing before being used. In general, Big Data sources can be effectively used as additional sources, benchmarks or proxies.


And last but not least, the same Big Data sets can be used in several contexts and for different purposes, e.g. Twitter data for mobility and public sentiment analysis.


We are now moving to the next phase of this project, where the aim is to produce a set of internationally comparable statistics based on Big Data before the end of 2015. 


If you are interested in being involved in this work, please contact the UNECE Statistical Division at [email protected]   

If you wish to subscribe to the UNECE Weekly newsletter, please send an email to:  [email protected]