The smartphone in your pocket is constantly collecting data on where and how you move. Your movements also leave information on the various cameras in our environment, travel card readers and payment terminals in shops.
Data accumulated in various databases could help city planners who are trying find ways of building safer cities that work better.
“However, spatial data from different sources are still spread out and exist in several formats, where their quality varies and the quantity is huge. The fragmented data must be combined into a form that we can use,” asserts Laura Ruotsalainen, an expert of navigation research.
She began her work as associate professor of spatiotemporal data analysis at the Helsinki Centre for Data Science, HiData in August. The perspective is that of sustainability science, particularly urban planning. Ruotsalainen’s duties include the development of algorithms to model measuring errors in location-based data and to aggregate the results.
“In an ideal world, our smartphones would be collecting correct data for navigation without the user even noticing or having to do anything. The device would be able to switch location methods seamlessly depending on which one is the most viable in each environment,” says Ruotsalainen.
For the average smartphone addict, this could mean that it would be possible to navigate from home to a meeting in another city, while directions for routes, different modes of transportation and the correct meeting room would automatically update on the phone. At the moment, this is only possible by using a number of different applications.
“This means that life will become safer and more comfortable without fanfare,” says Ruotsalainen.
But we do not live in an ideal world. For example, satellite navigation, popular as it is, is never perfect. It works poorly in environments with high buildings and is useless indoors.
The small, cheap sensors in our smartphones are currently unable to produce precise location data. Individual measuring results must also be cleaned before they can be used, which requires the use of error models created through statistical analysis. After that, the data must be combined through data fusion.
Machine learning is also necessary to work on the data. The use of machine learning is also a significant part of the field of Ruotsalainen’s associate professorship.
“Machine learning can help us recognise movement and model the ways in which people move. We can teach the system to notice that when these kinds of measurements start showing up from the smartphone, the person is walking or running. This will help us improve the location data and produce relevant information about human behaviour in a variety of different places and settings,” Ruotsalainen explains.
Map providers, transportation authorities and tech companies should also be encouraged to develop services that could form a seamless, universal location ecosystem. One of Ruotsalainen’s duties is to generate the technological foundation for such an ecosystem.
“We need a variety of technologies which could be combined through different forms of data fusion as the situation warrants. For example, a self-driving car should be able to change its navigation method on the fly, according to the environment and traffic situation,” states Ruotsalainen.
Security cameras generate urban data
According to Ruotsalainen, location-based services should be taking advantage of image data more frequently. Photos could help urban planners better understand how humans move through a city. One way of doing this would be to use material from security cameras, but this is questionable in terms of privacy.
Ruotsalainen emphasises that data protection is a primary concern.
“People must be able to decide how their data are used. Images from security cameras should only be used for purposes that have been explicitly agreed upon. It must be possible to erase faces, or to only extract data which can be used without identifying people.”
Ruotsalainen points out that this is currently difficult in practice, and not just in terms of face recognition. Corporations such as Apple and Google, which operate with huge amounts of location data, claim that they do not identify information to specific users. In reality, it would be easy to deduce a person’s home or place of work based on repeated location pings.
“We must keep ethics at the core of development. Our goal is an easier life, not surreptitious surveillance.”
Introducing people behind HiData
This series will introduce new professors in the tenure track system of the University of Helsinki working at the Helsinki Centre for Data Science.
Other parts of the series:
Keijo Heljanko, professor of parallel and distributed data science: Increasing masses of data may leave computers behind and cause an energy crisis
Kai Puolamäki, associate professor of data science and atmospheric sciences: Data science interprets atmospheric particles and helps find the cleanest urban routes – if we know what to ask computers
Nikolaj Tatti, associate professor of privacy-aware and secure data science: Data science may soon expose fake news
Antti Honkela, associate professor of data science - machine learning and AI: Everyone has their secrets – machine learning needs to respect privacy
Dorota Głowacka, assistant professor of machine learning and data science: Future search engines will help users find information they don’t even know they are looking for