Intelligently delivering OU content in a wider variety of formats to suit the particular needs of our students at any given time.
One of the benefits of smart technology is that it generally involves an element of awareness of the user's location, behaviour, and preferences, built up over time and uses smart assistants to provide suggestions and tips. For the purposes of this strand of the project we are looking into whether it is possible to utilise these technologies to provide content in alternative forms, delivered on the recommendation of the AI according to the user's preferences and the situation they are currently in.
Such an example could be whether it is possible for Alexa or Google Assistant to promote an audio version of a piece of content, rather than a textual one, for the student to listen to while they are on the move, perhaps while travelling to work? In such an example, the default offering in the VLE (assuming the student logs in and navigates to it) would be a piece of text explaining a particular theme, that would usually be read over a few pages in the module website, necessitating use of a screen and interaction with a mouse or touch-based interaction.
Instead, in the smart version, the University's AI-based system would establish that the user is on the move, that they are actively logged in on, say, a mobile phone and that they are using headphones. It would recognise that this is part of a daily pattern of behaviour (where the student travels daily to work) and that roughly takes about thirty minutes, during which time they usually listen to audio files or perhaps a podcast on another app on their phone.
The AI would review the available options for the content to deliver, establishing the week and point of study the student is at in the module based on their previous study sessions (established from login times and dates and progress) and then recommend to the user that an audio version of the current text has been created automatically for them to listen to should they wish. Alternatively the AI assistant could read the text to them using a text-to-speech engine.
The student could utilise this alternative option while travelling, and regardless of whether they finish the whole piece of content or only get partially through it, the assistant would record the point at which they finished listening and make this the default resumption point when the student returns to study. That resumption point could be another suggested study session by the assistant, for example at lunchtime or when the student gets home, or it could be a standard login to the VLE using a browser.
A key benefit here is that the student can resume easily from where they left off, having been aided by an automated, intelligent system in using an alternative format of the learning content, whereas they may not otherwise have utilised that time for study if they would have needed to read the text on a screen.
There are clearly multiple possibilities in such a system, but the basic concept remains the same - the smart system either creates or selects an alternative piece of learning content from the current default option in the module website, and based on a built-up knowledge of the student makes a recommendation that can either be taken up or ignored. Even if ignored, the system learns from that choice and uses it to make better future recommendations, so that it is always learning and improving. The student benefits from the suggestions by being able to make use of small pockets of time that may otherwise have gone by without being used.
There is also an accessible element to this strand, in that if the system can learn and build up a set of recommendations according to the user's unique status, situation, and needs, then it can take into account their accessibility requirements and provide a level of support that we currently do not. The more the system is used, the better it gets, not just for that particular student, but by using anonymised data sets, it can learn on how to better support students with similar needs in the future.
New students to the University could therefore benefit from past students activities through such recommendations, which can always be reviewed by the various faculty, production, design, and accessibility teams at the University at any time to ensure they are doing what they should.
Alternatively the student can choose not to use any of the above automated processes and simply use the devices to access their designated, chosen preference up front, for example by pre-selecting 'always use audio instead of text' and the system provides that option at all times instead of the text default when the student logs in. They could even turn off data collection, if they should prefer to keep their activities totally private.
The aim here, as with all strands of the Smart Tech project, is to test and evaluate what is currently possible with the technology and software and to establish what we could potentially do with minor amends and small investments, as opposed to redesigning the whole student experience or spending millions developing our own versions of pre-existing technologies. The benefits will vary greatly depending on the findings of our testing, but there is already evidence from previous academic and market research to demonstrate that we should be looking into such a system.
Work on this strand is scheduled to begin in February. Stay tuned here for regular updates.
The Smart Tech project was born out of the need to understand more fully the potential benefits of smart device usage on student success and to assess whether there are changes that we can make to the production and delivery of learning materials to assist in this aim.Discover more about the project