Page 320 - Special Topic Session (STS) - Volume 4
P. 320
STS637 Ronald W. J.
differences in weekday and weekend population, or inform about where
people are going on vacation domestically and how many foreigners are
visiting your country. All of those data are important to take decisions on
public transportation, housing and infrastructure.
Figure 3. Flow chart of the processing of mobile phone records
The statistical office of Indonesia with support of Positium use mobile
10
phone positioning data to estimate domestic tourism , internal migration,
commuting patterns and population density. The positioning data (monitored
over an extended period of time) will show a pattern about where a person
lives, where he or she works and what some of her or his regular activities are.
As illustrated in Figure 3, a lot of work is being done to pre-process data to
take out “noise” and subsequently to run algorithms, which turn data records
into statistics.
4. Conclusion
11
Tom Smith, Director of ONS Data Science Campus, presented recently
about “Data Science as a Team Sport” at the 5th International Conference on
Big Data for official statistics in Kigali, Rwanda. He emphasized that
partnerships and knowledge exchange are crucial elements for a successful
application of Data Science. In the UK, a Government Data Science partnership
was created to solve complex business problems using a combination of
domain expertise, coding knowledge, machine learning and statistics skills on
large and varied datasets.
Statistical institutes have started creating data innovation centers to
experiment with Data Science and Big Data sources. Besides the Data Science
10 See http://www.15th-tourism-stats-forum.com/pdf/Papers/S3/3 2 Indonesia's Experience
of using Signaling MPD for Official Tourism Statistics.pdf
11 See Conference agenda, Day 2, session at “Data Science And Capacity Development In
Official Statistics”, https://unstats.un.org/unsd/bigdata/conferences/2019/default.asp
309 | I S I W S C 2 0 1 9