Using natural language processing, researchers from Penn State developed a program that offers instant essay feedback

byHolden Galusha

| 2 min read

Share this Article

Listen with Speechify

0:00

2:00

Two recent papers, one published by the International Society for the Learning Sciences (ISLS) and the other in the Proceedings of the International Conference on Artificial Intelligence in Education (AIED), share the results of using a form of artificial intelligence called natural language processing (NLP) to provide immediate feedback on student essays and build summaries that allow teachers to more easily evaluate student comprehension. The studies were led by Rebecca Passonneau, PhD, professor of computer science and engineering at Penn State University. The paper published by ISLS reported the results of using an NLP program to assess student essays. In the AIED paper, the authors detail how they created the NLP assessment program by modifying existing software.

The team extended an NLP tool called PyrEval, which Passonneau and fellow researchers developed a few years ago, to assess the ideas presented by students in their essays. According to its GitHub page, PyrEval is a tool that automates the Pyramid method of evaluating content summaries. Passonneau and her collaborators developed the manual version of the Pyramid method in 2004 before automating it with PyrEval years later.

The extended version of PyrEval, dubbed PyrEval-CR, "can provide middle school students immediate feedback on their science essays," Passonneau said. This reduces the teacher's workload and eliminates the turnaround time between a student submitting a draft and receiving edits. "Simultaneously, the software generates a summary report on topics or ideas present in the essays from one or more classrooms, so teachers can quickly determine if students have genuinely understood a science lesson,” she added.

Exploring the applications of NLP in classroom settings has been explored previously by other groups. In 2012, researchers evaluated how NLP could be used as a sort of digital tutor to provide students with "writing strategy training." Additionally, a 2018 study examined how NLP could augment project-based learning curriculums.

The field of AI is advancing rapidly. As more studies and products are released, it's possible that these technologies will be tightly integrated within education systems worldwide.

About the Author

Holden Galusha

Holden Galusha is the associate editor for Lab Manager. He was a freelance contributing writer for Lab Manager before being invited to join the team full-time. Previously, he was the content manager for lab equipment vendor New Life Scientific, Inc., where he wrote articles covering lab instrumentation and processes. Additionally, Holden has an associate of science degree in web/computer programming from Rhodes State College, which informs his content regarding laboratory software, cybersecurity, and other related topics. In 2024, he was one of just three journalists awarded the Young Leaders Scholarship by the American Society of Business Publication Editors. You can reach Holden at hgalusha@labmanager.com.