DataSpark analysis provides insight into emergency calls
Hotline phone services – from the emergency 999 and 995 numbers to the weather line – are a vital source of security and information for Singaporeans. Now data science is being used to gain a deeper insight into how and when these phone numbers are used.
Experts from DataSpark – SingTel’s data analytics subsidiary – are analysing phone data to better understand the profile and behaviours of those calling the numbers with the aim of improving delivery of these essential services.
DataSpark uses the rich data provided by SingTel’s 5.5 million mobile customers. It gathers users’ location data as they move through the SingTel network and analyses this information to gain insights into their lifestyles, interests and movements.
All data used by DataSpark is encrypted, anonymised and aggregated to be compliant with SingTel’s data governance framework and with Singapore’s Personal Data Privacy Act.
The hotline services used in the analysis were:
- Police Emergency – 999
- Emergency Ambulance and Fire – 995
- Weather – 65427788
- Police Hotline – 1800 2550 000
- Dengue Hotline- 1800 9336 483
- City Gas – 1800 7521 800
- Power Grid -1800 7788 888
- Non-emergency ambulance – 1777
- Emergency Road Service – 67489911
Understanding the reaction to an emergency
The hotline project is in its early stages but according to DataSpark data scientist, Keen Dang, it has already provided some valuable insights. “We found that we can understand people’s reaction to specific events and then predict the behaviour of the public in the event of, for example, an earthquake,” he said.
“City planners can use DataSpark analysis to identify hotspots. We can even predict the number of crimes that are likely to occur in a particular area.”
An aid to facilities planning
He added: “If a lot of people are calling emergency services from the same area, we can look at the analysis to determine if there needs to be better facilities in that area.”
Dang said that only an initial study had been undertaken, but the results had been very promising. “I think this should be a very good tool if we can have further cooperation from different agencies and get some different data sources in addition to the telecommunications data, so we can get a better idea of what is going on. We plan to cooperate with other agencies that are interested, to share some of the results and do some more analysis.”
Not just for emergencies
This type of data analytics can be useful in other areas aside from emergency planning. According to Dang, “The technology can be applied to many services. For example, we could use it understand the behaviour of people calling taxis.”