Empowering Learning and well-being in the City of Helsinki using data and AI

A global leader in digital transformation of Education

In 2016, Helsinki was the first capital to have a digitalisation strategy for Education. Helsinki has been developing digital learning and learning analytics to make the city the most impactful place for learning in the whole world. The strategy involves viewing the entire city as a learning environment.

“We are aiming for systemic change,” emphasises Pasi Silander, “to do this, we can’t just change one piece of the jigsaw. We need to change all the pieces, all at the same time. It isn’t just about the technology or the operating culture or the pedagogy—it is all of those things plus the leadership strategy and the structure of the school. Everything has to be tackled all at once if the pieces are going to fit together and we are going to see a different picture.”

The first step on the city’s journey with data and AI was to identify the pedagogical use cases where   technology can support the “well-learning” vision. Treating each student as an individual and an equal learner has been critical to the City of Helsinki’s approach. This has a practical benefit in terms of facilitating teachers’ work and improving learning outcomes, as Pasi Silander explains, “teachers have maybe 25 to 30 students in a classroom and all the students are individuals with individual needs. Even though our teachers are Super Ladies and Super Men, they can’t really personalise each of the student’s learning processes.”  We will have to provide tools like AI or learning analytics for teachers to personalise students’ learning.

Cloud computing as enabler 

The team explored the ways in which AI and data analytics could enhance teaching and learning. Based on those use cases, the team had defined the technological and data architectures necessary to fulfil its vision. To achieve this, the city chose to work with Microsoft Azure to build an AI Hub.

Elias Vakkuri, Consulting Technology Architect for the City of Helsinki, outlines three key reasons for choosing Microsoft Azure: first, the range of integrated tools; second, the security and privacy credentials of Microsoft Azure; third, the collaboration and support that Microsoft provides. 

“Microsoft Azure offers us a number of ready-made AI components and other components which are very well integrated with each other,” says Elias Vakkuri. “This is important because we definitely see Microsoft Azure as the brains of our AI Hub—not just a storage or hosting solution.” 

Pedagogy merging with technology

Key use cases the development team chose to target included: dashboards for pedagogical leadership, on-time completion of studies, well-being and socio-emotional skills, scaffolding and automated feedback, adaptive learning processes and materials, and big data for research.

The City of Helsinki Education Division, with 100 000 students and 14 000 teachers and staff, is big enough to produce the sizeable datasets that are required to undertake big data. The City of Helsinki Education Division is a founding member of Microsoft’s Leaders in Digital Transformation of Education (LDTE) program and the Open Education Analytics Accelerator, formerly known as Project Constellation.

Through the collaboration, the City of Helsinki Education Division was able to utilise many resources, including developing its digitalisation strategy using the Education Transformation Framework. 

“Experts at Microsoft understand where we are headed. We couldn’t have built up our digitalisation strategy and benchmark what we are developing against other education systems without Microsoft’s unique knowledge and help.” 

Making the invisible visible

“With data, we can make the invisible visible,” says Pasi Silander. “A lot of things have been invisible in our global education systems but now, when we collect and analyse the data, we can make these things visible and make better decisions for students’ future.”

Technology and data alone are not the full story. The technology investments only make sense if the new technologies don’t just replicate existing processes but enable entirely new functions and innovations. Educators must be able to do new things with technology—beyond what one can do with pen and paper. AI and learning analytics have the potential to fundamentally change pedagogies—and that requires a big change in mindset. 

“Providing the technology or data is not enough,” confirms Pasi Silander. “We will have to achieve real pedagogical added value. We need to also educate policy makers and educators about how to lead with data and AI and to create a new data-driven culture.”

When it comes to data and AI, a global network of organisations partnering via Open Education Analytics (OEA) has been extremely helpful, along with the involvement of Microsoft product teams and experts. “It has been very important to have a dialogue with others on the same journey. It is very difficult or even impossible to find this kind of community within one country, to be one rebel without others to support you,” Pasi Silander outlines. These collaborations have contributed to the development of the Open Education Analytics resources that are now available for any education institution to use globally. 

The importance of ethics in AI

Data ethics and privacy are also cornerstones of the City of Helsinki Education Division’s work. 

“Because we are dealing with the children’s future and their sensitive data, we really emphasise    privacy and the ethics of AI,” states Pasi Silander.  

“Our key principle is that we don’t use learning analytics to classify students,” he notes. “All students are treated as individual learners. We must avoid potential self-fulfilling prophecies that could narrow students’ learning opportunities or stigmatise students.”

“Ethics is an end-to-end process. It starts with policymaking, then decision making, then software design, then designing what data to use, then training algorithms, then checking how end users are using the data and the results. The ethics governance for the whole end-to-end process is an essential part when building Learning Analytics and AI in Education,” outlines Pasi Silander.

This approach is in alignment with Microsoft’s responsible AI principles, a comprehensive model which covers fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability.

The future of education in the City of Helsinki 

“Development of advanced learning analytics and AI is a never-ending process,” concludes Pasi Silander. “We can’t predict the future. Society is continuously changing. Our models might work now, but when the population changes, these methods will need to be monitored and changed.”

While the strategies and the algorithms might have to continually evolve, the secure and robust platform provided by Microsoft Azure will continue to be the hub of data analytics and AI activity. Elias Vakkuri says, “We have received a lot of good commitment and lots of technical help from Microsoft and we are really looking forward to continuing that collaboration going forward.”

“A school is not only a building, it is a culture of competence development in a global learning environment,” says Pasi Silander. “We can’t teach kids to compete with computers and AI, so we need to think about what are the human skills we will need in the future? That’s the critical thinking, creativity, the soft skills, the socio-emotional skills, and that holistic approach to learning will need to be even more personalised.”

“To develop AI and learning analytics is a global opportunity for us. It is something we cannot do in isolation. We need companies like Microsoft and other stakeholders like cities and governments around the world to develop a bright future for our kids.” 

Interviewed June 2020