What role is there for big data in education?

Education at first glance seems to be an area where big data is yet to find the right fit. Most certainly there is the usual academic scoring and dare I say it, ranking, but what else is there for big data in education? Well we can also aggregate generic and public data in order to obtain a better view of where teaching and other educational resourcing needs to be applied.

That further data might be cultural, demographic, geo-location, gender, and many more attributes, all of which when considered along with the core academic data, certainly adds context. However, is this really shedding light on what we are trying to achieve in education. And when I say “we”, I don’t just mean a school or college, I mean the wider set of people who have an interest in students’ education being parents, teachers, policy planners, curriculum advisors, employers, and others.

When all these interested people are taken into account, how can big data help? Clearly, the issue becomes much more complex than simply accounting for academic achievement and looking at where help may be needed in the system and at the student level. Consequently, what is happening in education recently is the development of programs that are now looking at student education more holistically.

Characteristics such as innovative thinking, social responsibility, leadership, communication, art appreciation, sporting ability, peer acceptance, visualisation skills, spatial awareness, resourcefulness, resilience, and many more, are now being considered as part of a holistic education approach. Being able to define what attributes contribute to each of these characteristics is difficult enough let alone measuring them! However, there may be an answer.

Modern developments in semantic processing by which data is linked in a relationship connection is facilitating an insightful view of students’ holistic development that previously was not achievable in any objective manner. Semantic processing solutions such as Latize Ulysses are able to use human-like assessment to link data points that otherwise would not seem to be connected or relevant. Machine learning assisted mapping techniques allow column definitions to be automatically mapped from raw data sources to semantic graphs, making data transformation and visualisation more efficient. When presented to users, the semantic graphs are highly consumable in terms of providing the required student insights.

With Ulysses, educators can connect to virtually any student academic scores, extracurricular activities, demographic statistics, attendance records, assessments, Learning Management Systems, and survey data, easily, thereby obtaining a holistic student view.  That source information could be in almost any form and will be mapped and linked according to the underlying principles perceived as relevant. With a dynamic and highly customizable dashboard, educators can define alerting rules to flag out students who may be at risk and make an early intervention to improve student and education outcomes.

Finally, according to the organisations capacity and resources, a range of intervention strategies can be incorporated into the solution for automatic recommendation by the system according to the surfaced results.