Researchers have designed an artificial intelligence (AI) model that is better able to predict how much students are learning in educational games.
The improved model makes use of an AI training concept called multi-task learning, and could be used to improve both instruction and learning outcomes.
Multi-task learning is an approach in which one model is asked to perform multiple tasks.
“In our case, we wanted the model to be able to predict whether a student would answer each question on a test correctly, based on the student’s behavior while playing an educational game called Crystal Island,” says Jonathan Rowe, co-author of a paper on the work and a research scientist in North Carolina State University’s Center for Educational Informatics (CEI).
“The standard approach for solving this problem looks only at overall test score, viewing the test as one task,” Rowe says. “In the context of our multi-task learning framework, the model has 17 tasks – because the test has 17 questions.”
The researchers had gameplay and testing data from 181 students. The AI could look at each student’s gameplay and at how each student answered Question 1 on the test. By identifying common behaviors of students who answered Question 1 correctly, and common behaviors of students who got Question 1 wrong, the AI could determine how a new student would answer Question 1.
This function is performed for every question at the same time; the gameplay being reviewed for a given student is the same, but the AI looks at that behavior in the context of Question 2, Question 3, and so on.
And this multi-task approach made a difference. The researchers found that the multi-task model was about 10 percent more accurate than other models that relied on conventional AI training methods.
“We envision this type of model being used in a couple of ways that can benefit students,” says Michael Geden, first author of the paper and a postdoctoral researcher at NC State. “It could be used to notify teachers when a student’s gameplay suggests the student may need additional instruction. It could also be used to facilitate adaptive gameplay features in the game itself. For example, altering a storyline in order to revisit the concepts that a student is struggling with.
“Psychology has long recognized that different questions have different values,” Geden says. “Our work here takes an interdisciplinary approach that marries this aspect of psychology with deep learning and machine learning approaches to AI.”
“This also opens the door to incorporating more complex modeling techniques into educational software – particularly educational software that adapts to the needs of the student,” says Andrew Emerson, co-author of the paper and a Ph.D. student at NC State.
The paper, “Predictive Student Modeling in Educational Games with Multi-Task Learning,” will be presented at the 34th AAAI Conference on Artificial Intelligence, being held Feb. 7-12 in New York, N.Y. The paper was co-authored by James Lester, Distinguished University Professor of Computer Science and director of CEI at NC State; and by Roger Azevedo of the University of Central Florida.
The Latest Updates from Bing News & Google News
Go deeper with Bing News on:
- How Uber Implements CI/CD Of Machine Learning Modelson July 8, 2021 at 3:30 am
The 'model deployment service' performs validation by making prediction calls to the 'candidate model' with ‘sampled data.’ ...
- Sound On: Audio Platform Khabri Makes News With Its Audio Courses On Government Job Examinationson July 6, 2021 at 6:34 am
Much like we would leisurely opt for listening to an audio book and multi-task our way through it; learning through audio for academics is taking shape. Mashable India talks to Mr. Pulkit Sharma, ...
- Using Machine Learning Methods to Predict Response to Immunotherapyon July 1, 2021 at 1:44 pm
“We considered these proxies as different tasks to be predicted by our machine learning models and used multi-task learning algorithms in order to learn all tasks jointly.” The researchers ...
- What Pandemic Parenting Can Teach Us About Leadershipon June 30, 2021 at 8:12 am
Second, learn how to serial-task (not multi-task). Third, manage a diverse team ... Talking to other busy WFH parents and learning what they were doing and why helped them navigate the unknown.
- Long Read: Blinkist co-founder and CEO Holger Seim on why learning should be a constanton June 25, 2021 at 7:56 pm
Holger Seim along with Sebastian Klein, Tobias Balling and Niklas Jansen founded the ‘millennial’ app in 2012.
Go deeper with Google Headlines on:
Go deeper with Bing News on:
AI predictive model
- Verge Genomics Announces Three-Year Collaboration With Lilly to Discover and Develop Novel Treatments Using Its AI-Driven All-in-Human Platformon July 8, 2021 at 3:30 am
Verge Genomics, a biotech company that has created an industry leading all-in-human, artificial-intelligence-powered drug discovery and development platform focused on therapies for serious genetic ...
- Deploying AI-powered modelling to reduce the cost of claimson July 7, 2021 at 9:02 pm
Over the next decade the claims process will be transformed, with what McKinsey and Company describes as transformative, next-generation capabilities. The authors of its April 2019 report, “Claims ...
- Strategies for Using Predictive Analytics, AI to Improve Careon July 7, 2021 at 10:00 am
How predictive analytics and artificial intelligence can improve the patient experience and improve the quality of care when incorporated into the healthcare system.
- Top 10 Ideas in Statistics That Have Powered the AI Revolutionon July 6, 2021 at 3:45 pm
Andrew Gelman, a statistics professor at Columbia, and Aki Vehtari, a computer science professor at Finland’s Aalto University, recently published a list of the most important statistical ideas in the ...
- Why sepsis prediction models fail: 4 things to knowon July 6, 2021 at 9:00 am
LinkedIn blog post, Ms. Russell, a senior clinical data engineer at Renton, Wash.-based Providence, gave four reasons that sepsis predictive models fail. Her article comes in response to a recent JAMA ...