Chung-Ang University researchers win five papers at top-tier data-mining conference CIKM 2026

Researchers from Chung-Ang University’s College of Software will present five papers at CIKM 2026, a leading conference covering data mining, information retrieval, knowledge management and databases.

The conference, the 35th ACM International Conference on Information and Knowledge Management, will take place in Rome in November. The university said its papers span several research areas, including knowledge graphs, large language models, data quality, generative models and recommendation systems.

Professor Kim Moo-chul’s team had two papers accepted. One proposes a method for incorporating morphological information from agglutinative languages such as Korean into language models. The paper received an oral presentation slot in the Full Research Papers Track. Another study examines how data-quality filters for training large language models can produce different evaluations depending on document structure.

“

The acceptance of five papers at CIKM 2026 demonstrates the solid research base and world-class competitiveness of our university in data mining.

Professor Ko Yoon-yong’s team will present PROBE-Web in the Demonstration Track. The project is an interactive evaluation system for analyzing and comparing knowledge-graph completion models from different perspectives.

Professor Jo Yun-sik’s team worked with content platform WATCHA to study bias in collaborative filtering and propose a new training method. Professor Lee Hwan-hee’s team, through a research internship with LG AI Research, developed a method for expanding story-focused creative-writing data across genres. The college dean, Park Sang-oh, described the five accepted papers as evidence of broad and internationally competitive research across the college.

“

We will continue to provide active support so that research achievements recognized on the global stage can continue.