Managed by Algorithms


In my last article I explored the growing use of cognitive services within the education sector. I stated that the use of machine learning within the education sector is based on a simple premise; namely that education institutions are essentially information-processing entities. If this is the case, then student and institutional data can be leveraged by cognitive service platforms to inform the myriad of algorithms that will come to inform the behaviour of students, teachers and support teams on the campus. As algorithms begin to play a greater role on the campus they will invariably augment the capabilities of teachers and support teams; and by doing so they will enable the institution to be better placed at providing on-demand services for students and thereby add value to every aspect of the student life cycle.

The use of algorithms to support our day-to-day affairs has been with us for many years. We are influenced by their behaviour whenever we use the sat-nav in our cars, whenever we browse or shop online or whenever we apply for a loan or insurance cover. However, the extensive use of algorithms to manage the myriad of services within a school, college or university is still in its infancy; especially when it comes to the management of support services for students and employees on the campus and the management of teaching, learning and assessment.

How will algorithms manifest themselves within an education setting?
Many of the routine tasks that are undertaken by students, teachers and support teams and the decisions that they make may be managed algorithmically. Here is a selection of potential use cases:

  1. The adaptive learning environment presents students with tutorials and assessment activities based on their track record on their courses. Online content and assessment activities adapt and contextualise to the needs of each student. Algorithms also determine the feedback that is presented to each student.
  2. The student home page or the institution's student app takes advantage of a suite of algorithms to determine which information, advice and guidance to present to each student - this reflects the current status of each student and the desired goals and aspirations of each student.
  3. The library management information system identifies classroom topics, the coursework or the assignments that students are currently doing and presents each student with a personalised set of resources that could be used to support their studies.
  4. Tutors and support teams are alerted to all the students who are at risk of under-performing with their studies or who may be at risk of leaving their studies. The algorithms or agents may also offer suggested actions to support the student. They may even carry out these actions on behalf of the tutors and student support teams.
  5. Class timetables are no longer static because they adapt and change according to the prevailing demand for classrooms on the campus.
  6. Part-time teaching staff are assigned algorithmically to classes according to anticipated demand.
  7. Management teams are presented with real-time insights regarding the performance of individual teams, departments and the campus as a whole.
  8. The school, college or university automates the communication of prompts or notifications with its students - these are bespoke to each student regardless about the size of the institution.
  9. Long form answers are automatically marked and graded. The feedback that students receive is contextualised and personalised according to the academic needs of each student.
  10. The content and course listings that are displayed on an institution's website are determined algorithmically; reflecting the needs and demands of potential students and that of the institution.

I am sure you can identify many more use cases for your school, college or university. The majority of these algorithms may be welcomed by students, teachers and support teams on the campus; because they remove many routine and repetitive tasks from their daily schedules.

I thought it would be useful to describe the nature, context and behaviour of these algorithms. Just as teachers and student support teams behave cooperatively to support the academic and pastoral well-being of their students; algorithms, cognitive agents or micro-services will also be required to cooperate with one another to support the needs of students, teachers and support teams on the campus. The performance of these micro-services can be managed iteratively; enabling institutions to swap, enhance and change them as and when required.

As algorithms mature they start to function and behave with a greater degree of autonomy. In this context they can be referred to as agents. Within the context of a school, college or university agents can be described as programs that observe the behaviour of students, teachers and support teams as they interact with the services on the campus. These agents undertake data mining activities which enable them to extract meaning and knowledge from the large datasets that are to be found in a modern education setting. The agents then direct or combine their activities to satisfy the needs of students, teachers, support teams and that of the wider institution. These agents will support every facet of life on the campus. As you can imagine, there could be many thousands of them - all working cooperatively to support everyone on the campus with day-to-day tasks and activities. The article entitled Learning Support Agents explores the role of these agents on the campus.

If we are going to trust these algorithms to work on behalf of the institution to help students, teachers and support teams they need to be appropriately designed, trained and managed. Everyone on the campus will need to consider how these algorithms will behave in a manner that supports the underlying ethos, ethics and principles of the wider institution. The design of algorithms or agents should not be taken likely; especially when the decisions that they make impacts so significantly on the individual at the receiving end. For example, agents could be used to filter student applications to courses; they could be used to direct students to particular outcomes on their courses; they could be used to assess student work and to offer feedback to the student; they could be used to deliver advice and guidance to students about future programmes of study; they could be used to cancel the offer of planned courses; and more. If algorithms or agents are going to be used wisely, appropriately and with sensitivity, institutions should consider the appointment of an ethics lead whose responsibility will be to monitor the use of these services on the campus; advising teams about the legal use of algorithms, protecting individuals from bias and to promote the positive and productive use of algorithms. These algorithms or agents need to behave in a manner that enables students to be in greater control of their studies, enabling them to make informed decisions, enabling them to succeed with their studies, and to be an active and productive member of the wider community on the campus.