Ka Tim Chu , a teacher and deputy principal of True Light College in Hong Kong, wanted to understand the emotions of his students during class. Now a new learning platform using AI technology can monitor children's emotions as they take online classes.
The software, called 4 Little Trees, was created by Find Solution AI, a Hong Kong-based startup. While the use of AI technology to recognize emotions in schools has raised concerns, its creator Viola Lam says it can make a virtual classroom as good as a real one, if not better.
Students work on tests and homework on the platform as part of their school curriculum. While they study, artificial intelligence measures the muscle tone in their faces through their computer or tablet's camera and identifies emotions such as happiness, sadness, anger, surprise and fear.
The system also tracks how long it takes students to answer questions. It records their marks and performance history. It generates reports on their strengths, weaknesses and motivation levels and predicts their grades. The program can be tailored to each student, targeting gaps in their knowledge and offering game-style tests designed to make learning more fun. Students perform 10% better on exams if they have learned to use 4 Little Trees, says Lam.
Lam, a former teacher, remembers discovering that some of her students had gaps only when she saw their test results.
The 4 Little Trees software was launched in 2017 with $5 million in funding to give teachers a better understanding of the challenges their students face. The number of schools using it in Hong Kong has grown from 34 to 83 in the past year. Prices range from $10 to $49 per student.
Lam says technology is particularly useful for teachers during the pandemic because it allows them to remotely monitor their students' emotions as they teach.
Chu believes the benefits of the technology will continue after the pandemicbecause it reduces the burden on administrators by creating and marking personalized assignments and tests. And unlike teachers, expression-reading AI can pay close attention to each student’s emotions, even in a large classroom.
But technology that tracks children's faces raises privacy concerns.
In China, artificial intelligence analyzing biometric data for surveillance purposes in schools and other places has sparked controversy.
Lam says 4 Little Trees records facial muscle data. This way, the AI interprets emotional expressions, but it doesn't record the students' faces.
Pascale Fung, director of the Artificial Intelligence Research Center at the Hong Kong University of Science and Technology, says “transparency” is key to preserving students’ privacy. She says developers should get parental consent to collect student data and then “explain where that data will go.”
Racial bias is also a serious issue for AI. Research shows that some emotion-analysis technologies struggle to identify the emotions of darker-skinned people, in part because the algorithm is trained to recognize emotions primarily from white faces.
Lam says she trains the AI with facial data that matches the demographics of students. So far, it has worked well in Hong Kong's predominantly Chinese society, but she acknowledges that more mixed communities could pose a bigger challenge for the software.
Lam said Find Solution AI’s emotion recognition works with an accuracy of 85 percent in Hong Kong. Fung says that algorithms with “very good settings” can correctly identify primary emotions, such as happiness and sadness, up to 90 percent of the time.
However, more complex emotions like annoyance, excitement or anxiety can be harder to identify. As AI improves, Lam hopes to develop applications for businesses, as well as schools, to better understand participants’ needs and increase engagement in online meetings and seminars.
