Even After Scoring 95 in Math, "I'm Unhappy"... AI Reveals Children's True Feelings Unknown to Teachers [AI Revolution](186)

Interview with Tebasoft CEO Oh Jeongseop
"Simspace" Adopted by 1,500 Schools
Analyzing Student Diaries to Detect 46 Emotional Signals

Gemini-generated image

"How about writing down your thoughts on how this evaluation went?"

After completing a mathematics unit test, an elementary school teacher asked the students to write a short reflection. The teacher had divided the class into groups of fast and slow learners, so the outcome seemed predictable. Students who were good at math would likely be satisfied, while those who struggled would probably write that it was difficult.

However, the results analyzed by Tebasoft’s artificial intelligence (AI) emotion analysis platform, Simseupeiseu, were different. Even among students who scored 95 points, there were notable signals of 'unhappiness' and 'stress.' Conversely, some students who did not perform well still showed stable emotional states. Only then did the teacher realize:

'The reasons I thought the students would struggle, and the reasons I thought others would be fine, were all based on adult standards.'

Simseupeiseu, which analyzes students’ short writings using AI to reveal their emotional states, brings to light subtle signals in the classroom that are easily overlooked. It provides a snapshot of the class’s overall emotional climate and visualizes emotional states such as stress, anxiety, and depression as data. On the 3rd, Oh Jeongseop, CEO of Tebasoft (pictured), told The Asia Business Daily, "The thing that surprises teachers the most is precisely this unexpected result."

On the 3rd, Oh Jeongseop, CEO of Tebasoft, is introducing the Simseupeiseu feature in an interview with The Asia Business Daily. Photo by Park Yujin

After completing a Ph.D. program in computer engineering and working at Samsung Thales and as a research professor at KAIST, CEO Oh founded Tebasoft in 2022. The core of the Simseupeiseu business model originated from the concerns of his wife, an elementary school teacher. Starting from the practical challenge of "not understanding the students’ feelings," they created a system where AI analyzes student diaries and presents the results on a dashboard that teachers can view at a glance.

In 2024, the introduction of AI digital textbooks brought rapid changes to the school environment. Basic infrastructure such as Chromebooks was put in place, and the use of edtech spread simultaneously. Amid these changes, Simseupeiseu visualized the emotional climate of entire classes using red, yellow, and green signals, quantifying stress, depression, and anxiety so teachers could understand the situation at a glance. The data revealed patterns such as students who scored 95 in math but still felt unhappy, and students who frequently expressed 'annoyance' eventually experiencing conflicts.

These nuanced emotions manifested in the classroom could not be fully captured by traditional emotion analysis methods. Recognizing the limitations of the initial 13 emotion categories, Simseupeiseu transitioned to a large language model (LLM)-based system and expanded its emotional framework to 46 categories. As a result, teachers were able to observe students’ conditions not simply as normal or abnormal, but as trends and changes over time.

This is also where Simseupeiseu differentiates itself from competing services. Rather than merely analyzing individual emotions, it focuses on capturing emotional trends within school and classroom groups. The "group emotion analysis" technology, which detects outlier emotional changes or signals of social isolation within a group, is a core patent held by Tebasoft.

The usage fee for Simseupeiseu is approximately 40,000 won per class per month, and it is currently used by 1,500 schools. Of these, 800 schools adopted the platform without cost to the schools themselves through a contract with the Seoul Metropolitan Office of Education, and starting this year, a contract with the Gwangju Metropolitan Office of Education will also launch.

Having been selected for the Ministry of SMEs and Startups’ Global Track technology startup support program at the end of 2025, Tebasoft’s next target is the overseas school market. Currently, Simseupeiseu supports eight languages, including Korean, English, Japanese, Chinese, and Spanish, and pilot programs are underway in Mongolia and Japan. Discussions for local adoption are also ongoing in Vietnam, Sri Lanka, and other countries through local partners.

Currently, about 260,000 emotional data points accumulated in Simseupeiseu are being used not only for model advancement but also as emotional indicators that can be applied in schools. CEO Oh explained, "There are many moments when teachers confirm with data what they only vaguely suspected," adding, "In classroom environments where it’s difficult to look after each individual child, our goal is to serve as a supporting tool."

This content was produced with the assistance of AI translation services.