Content metadata: the 'why' and 'what' of learning objectives tagging and automated labelling
Companies that want to keep up with market developments require well-organised metadata at a very granular level. They need to embrace automated labelling to be ready for the future. But labels and metadata are used at various levels, which means it's not easy to see the forest for the trees.
In this blog series, we focus on content metadata. Now that we've discussed the CEFR, keyword extraction, and topic classification, let's have a look at learning objectives tagging.
A brief explanation
In our view, learning objectives tagging is a form of topic classification. But it has a specific taxonomy: a curriculum, or a structured set of learning goals.
As you can tackle labelling at different levels, you're dealing with a hierarchical taxonomy. For example, you can label subjects, but you can also take it one step further and label the topics that fall under certain subjects.
Why learning objectives tagging is useful
People tend to learn at their own pace, which means every student is at a specific point in the curriculum. In the past, fast learners had to wait weeks or months for their peers to catch up. But learning objectives tagging has paved the way for personalising teaching materials. Once a student has completed a certain part of the curriculum, you can easily have them move on with the next part. After all, you have labelled the curriculum meticulously, which allows you to customise a student's learning path.
Publishers can use learning objectives tagging to analyse their teaching materials. Does your curriculum cover all relevant topics? Have you created teaching materials for all learning goals in the national curriculum? You'll answer such questions much faster if you've embraced learning objectives tagging. On top of that, it's easier to curate and offer materials from third parties. And in today's world of content curation, that's invaluable to every publisher!
How to create automated labels
To be successful, an AI model will require labelled content and the associated taxonomy. Keyword analysis can help incorporate a taxonomy into the model. At Edia, we have a keyword extractor analyse a curriculum to understand what it's about and identify potential labels related to a certain topic. That is how we train an AI model to classify learning objectives.
The benefits of automation
Automation will allow educational publishers to add materials to their existing collection faster. Furthermore, it will be easy for them to spot (and fill) the gaps in their offer.
Ultimately, learning objectives tagging will open the door to new business models and revenue streams: publishers can take on the teacher's role and provide personalised learning materials.
In the past few weeks, we've discussed the content metadata we deem important in terms of automated labelling. Of course, there are many other types of labels. So, where to go from here? We'll discuss that in our next blog post, which will be the last one in this series.