Soonchunhyang University AI and Big Data Students Win Grand Prize at ETRI AI Paper Competition

Predicting Sleep Quality with Wearable Data
Winning the Minister of Science and ICT Award Among 370 Teams

Students from the Department of AI and Big Data at Soonchunhyang University won the Grand Prize (Minister of Science and ICT Award) at the 4th Human Understanding Artificial Intelligence Paper Competition, organized by the Electronics and Telecommunications Research Institute (ETRI).

They received the highest honor among 370 teams nationwide for their technology that predicts sleep quality using wearable data.

According to Soonchunhyang University on October 22, the competition was co-hosted by the Ministry of Science and ICT, the National Research Council of Science & Technology (NST), and ETRI, and operated by Dacon.

The event, held under the theme "Predicting Sleep Quality and Status Using Lifelog Data," ran for six months and attracted a total of 370 teams (1,034 participants).

Lee Gwangseop (senior) and Choi Seongmin (junior) from Soonchunhyang University received the Grand Prize for their paper, "Explainable Time Series Feature Engineering for Sleep Quality Prediction Using Lifelog Data."

The two students developed an artificial intelligence model that predicts sleep status by analyzing wearable sensor data.

They enhanced accuracy with an ensemble technique combining LightGBM and CatBoost, and visualized factors affecting sleep quality through SHAP analysis.

Through this, they demonstrated the potential of data-driven healthcare services.

Lee Gwangseop said, "I wanted to create not just simple records, but predictive health indicators," adding, "Based on this experience, I will take on the challenge of developing healthcare AI."

Professor Woo Jiyoung, the supervising professor, evaluated the research as "a meaningful study that scientifically interprets human health status using life data."

Jeon Changwan, head of the SW-Centered University Project Group (Vice President for Research and Industry-Academia Cooperation), stated, "The students demonstrated a high level of research by designing feature engineering themselves," and added, "We will continue to actively foster problem-solving SW convergence talent."

This achievement is part of Soonchunhyang University's SW-Centered University Project (supported by the Ministry of Science and ICT and the Institute for Information & Communications Technology Planning & Evaluation), and the university is striving to nurture practical AI and software talent with annual support of 2 billion won.