- Personalising the student learning experience
- Improving language and literacy skills
- Using data analytics to identify struggling students
- Gathering behavioural data to uncover student potential
- Automating tasks and increasing educators’ efficiency
Education and advancements in technology often go hand in hand. It should come as no surprise, then, that artificial intelligence, a relatively new development in technology, is already making an impact on the sector. With its unmatched ability to collect and analyse data, automate mundane tasks, and personalise learning experiences, AI can make the lives of teachers and students alike easier.
Personalising the student learning experience
One of the biggest problems with traditional teaching methods is that they fail to take into account that each student has unique learning needs and requirements, which can reflect negatively on their learning outcomes. Thankfully, AI can offer a helping hand in this regard. Teach to One 360 is an AI-enabled, personalised maths-learning system developed by Joel Rose and Cris Rush. Rose, a former teacher, had noticed that there was great discrepancy in maths skills among his students, even in the same class; while some were advanced by two grades, others were far behind. Rose realised that the main reason behind this lay in the outdated, one-size-fits-all teaching model that most schools employ. Teach to One 360 was, therefore, built around the concept of blended learning, offering personalised lessons to students specifically tailored to their needs and set of skills.
Teach to One 360’s algorithms and machine learning play a crucial role in determining a student’s maths level and, consequently, the appropriate lesson plan and teaching method. First-time users take a 90-minute standardised test measuring maths skills and a further 60-minute diagnostic test that determines their strengths and areas that need improvement. Teach to One 360’s versatility comes into play when the algorithm selects the appropriate modality (teaching method) for the day. Some kids might get a personalised one-on-one lesson from a tutor, some will work in a group with other students, while others will work independently using online interactive programmes and games. Once the lesson for that day is over, the program tests the students with a five-question quiz to track their progress and determines the next day’s lesson.
Another example of an AI-based personalisation tool comes from Carnegie Learning, an EdTech and curriculum solutions provider, which has earned a reputation for its powerful AI-enabled maths learning software for secondary schools. Its award-winning platform, MATHIa, is a one-to-one, virtual maths coach that provides students with real-time feedback and assessment. MATHIa essentially guides students step-by-step through the learning process, adjusting to their levels of knowledge and tracking their progress. It is also capable of predicting how far students will progress by the end of the year. Another feature that is particularly helpful for teachers is the ability to notify them when students are idle or need extra support. Carnegie Learning also offers solutions in ELA and applied sciences.
There are even AIs that offer complete courses. Squirrel AI is one of EdTech companies that offer students personalised online tuition at home. Using its algorithms, it can assess a student’s knowledge and skills, and create fully personalised programmes to fill knowledge gaps. To ensure efficiency, student learning progress is monitored in real time, so Squirrel AI can always adjust their learning strategy if needed. Since it actively looks at each student’s performance, Squirrel AI can predict how much time an individual pupil might spend on a specific knowledge point. For example, if a student has limited time to study, it can identify which knowledge points should not be studied based on their importance and the time it takes to digest them. “During our work, we found that students could be so different from one another,” says Derek Li Haoyang, chairman and founder of Squirrel AI. “For one particular knowledge point, one student could grasp it in seven minutes, but it might take at least 90 minutes for another.”
Improving language and literacy skills
Language learning is a particularly fruitful field for AI, considering the abilities of natural language processing and machine learning. Blue Canoe is a mobile app that aims to help users practice their spoken English to sound more confident and clear. uses the Color Vowel System methodology, whereby each vowel is assigned a colour and placed within a chart that reflects the shape of the mouth when the sound of the vowel is produced. This method is well-known and has been used in more than 10,000 classrooms over the world. Combining it with speech recognition and AI, Blue Canoe creates a virtual AI teacher for each user that provides immediate and helpful feedback. The app hosts interactive lessons, quizzes, personalised learning plans for premium users, and even a card game for learning vocabulary.
Still, learning with an instructor has obvious advantages over self-learning, and some platforms, like Ponddy Education, combine AI and human instructors to help their users master a language. Ponddy uses machine learning and natural language processing to create personalised language-learning services and study materials for Chinese learners. They have patented an AI-based technology called the Affinity Knowledge Learning System, which categorises words that often appear together and have shared properties. Ponddy also offers an online learning platform called Smart Textbooks, where teachers can create and customise curricula around thematic learning modules that come with text, vocabulary, grammar, and gamified tools. They can also track their students’ progress with real-time analytics.
AI can also help visually impaired students, who usually only read Braille. OrCam’s MyEye, for instance, allows the visually impaired to read books, distinguish products and brands, and recognise faces. It’s a small wearable device consisting of a camera and audio piece that can be attached to any pair of eyeglasses, and a base processing unit the size of a mobile phone. The device activates when the user clicks on a trigger button or simply points to a text with their finger. Once the camera and base unit process the text, the audio piece relays the information to the user. Seeing AI is another AI-powered app that helps visually impaired children see and read by analysing visual information in a photograph and describing it to them. And children with dyslexia can, apart from traditional exercises and methods, try practising with software that reads text out loud on a smartphone.
Using data analytics to identify struggling students
Today’s educational institutions generate vast amounts of data through their campus networks, infrastructure, applications, servers, learning management systems, and end-user devices. When leveraged properly, this data can provide invaluable insights into the problems, trends, and patterns that are being felt across the sector. The Catholic Diocese of Maitland-Newcastle recently joined forces with Microsoft to develop an advanced analytics and data science platform that can help identify struggling students, as well as those on the opposite end of the spectrum, by analysing data on their education history.
Each student is assigned a set of attributes, including classes, teachers, attendance, academic performance, and previous schools, which are then put through the predictive model to better understand student growth and offer assistance where needed. “We put a face on a student. So I can look at every student and see every one, every individual that touched that student at a certain point in time. We understand the student from his or her context. And by doing that we are able to have that targeted intervention,” explains Zanné van Wyk, the group’s head of data and analytics.
The social learning platform Piazza recently unveiled a new insights and analytics tool named Student Participation Export Tool, which enables educators to obtain valuable insights into student participation and engagement and identify areas where they may be struggling. “As classes around the world continue in a virtual environment, many students will struggle with the feeling of isolation, making it harder to stay engaged in their courses,” says Pooja Sankar, founder and CEO of Piazza. “We’re excited to provide instructors with a holistic view of which students are struggling and on what topics. Our hope is that this data empowers instructors to make informed decisions to reach out to students who need help.”
Gathering behavioural data to uncover student potential
However, data gathering and analysis can go a step further than simply deriving descriptive data from students. After all, a student is not a series of numbers that denote certain characteristics, but a dynamic being who is constantly interacting with their environment. The potential behind behavioural data, as opposed to descriptive data, is often underappreciated. This is why KnackApp, a California-based applications provider, has been designing games that also serve to inform us about people’s behaviours.
Specifically, KnackApp designs game-like applications that gather information pertaining to what are called ‘microbehaviours’, which relate to how the person functions around the app – do they repeat mistakes or learn? Do they experiment with new solutions? Of course, tracking these sorts of behavioural patterns can tell us a lot about how students learn and what sorts of methodologies may be more effective than others. Guy Halfteck, KnackApp’s founder and CEO, claims that these apps can start returning results after just 10 minutes of gameplay, providing a “powerful indication of your human operating system.” KnackApp’s games are meant to be able to “tease out, to measure and identify and discover those intangibles that tell us about the hidden talent, hidden abilities, hidden potential for success for that person.
The Big Picture Learning school in Italy is part of an international network of schools that encourage and embody innovation in education – a total of more than 300 schools across 60 countries or more. The school has been using the Knack solution to better engage its students. “Using Knack games to discover students’ hidden career potential enables Big Picture Learning to set its students onto personalized educational paths”, says the school’s international director Fabio Pirola. “We’ve been using Knack for the past three years to help our students gain self awareness, discover their unique talents, and navigate their future career path. Knack has been very useful, enabling our educators to transform student engagement and career planning while reducing school dropout.”
Automating tasks and increasing educators’ efficiency
AI technology can also help schools automate certain tasks, allowing their staff to focus on more important matters. With more and more students having to study and take exams online at home, schools encountered the problem of keeping invigilation fair and efficient. Clearly, it is impossible for a single teacher to monitor a class of 20, let alone 100 students, and prevent them from cheating. This is where AI can step in. Indian-based solution Eklavvya automates the online invigilation process. It uses a camera for face detection and live video streaming, and it can also interact with remote students using live chat.
Student assessments are another area where AI can be of help. A type of intelligent tutoring system, US-based Cognii uses natural language processing to provide real-time assessment of open-response answers provided by the student. It analyses the syntax, deeper semantics, and concept hierarchy (what concepts are present and which are absent in the student answer) to evaluate writing and offer immediate feedback. Cognii claims to be 96 per cent as accurate as humans in evaluating short essays. It can also work as a tutoring assistant, helping students improve reading comprehension, study STEAM, and social sciences. For educators, Cognii offers reports and graphical overviews of knowledge gaps and concept mastery at an individual and group levels.
What Cognii aims to do is provide an assessment that goes beyond multiple-choice questions. Open-response prompts challenge students to think critically, rather than guess or memorise a simple answer. The chatbot relies on data mining and machine learning to keep improving its responses and adaptiveness to student writing styles. Important to note is the fact that Cognii works with content provided by teachers to assess student answers and provide customised assessment tools. Teachers can update and edit questions with the correct responses and concepts, which Cognii uses to evaluate students’ answers.
As shown by the examples above, the potential of AI in education is truly limitless. When properly used, AI can make the learning process easier and more efficient. Its ability to automate certain activities and provide real-time feedback to both educators and learners will become indispensable as schools continue adopting new technologies and reorganising their resources. Despite some apparent drawbacks, the benefits offered by AI ensure that its adoption within secondary education will continue to grow in the upcoming years and will play a major part in educating the younger generations.