Back to the newsroom

News

Wicked Storm granted a patent for its xAPI learning anomaly detection system

Patent certificate No. 10-2883249: learning anomaly detection system based on xAPI learning behavior data
  • Granted Patent No. 10-2883249 for a “learning anomaly detection system based on xAPI learning behavior data”
  • Following the LLM-based learner profiling method (No. 10-2767110) granted in February, both AI patents are now granted
  • Applied to AI anomaly detection in Lecognizer: AI finds signals, and administrators and instructors judge

EdTech company Wicked Storm (CEO Jeongjun Lee) announced on November 4 that it has been granted a patent (No. 10-2883249) for a “learning anomaly detection system based on xAPI learning behavior data.” The invention was filed on August 20, 2024, and was granted on the same day after examination. The patent holder is Wicked Storm Inc.

xAPI is a global standard that records learning activities as standard statements made up of an actor, a verb, and an object. Every action that occurs during learning is stored as data in the same format. The patent covers a system that analyzes the xAPI learning behavior data collected in this way to detect anomalies that deviate from typical learning patterns.

Scores or result reports alone make it hard to know what difficulties arose during learning. Learning anomaly detection finds unusual patterns left in learning process data and points first to where attention is needed. This forms a foundation that supports the reliability of the collected learning data and the quality of analysis.

The technology forms the core of AI anomaly detection in Lecognizer, Wicked Storm’s learning data repository built on global standards. Lecognizer collects and stores learning data from all teaching, learning, and operational activities using global standards, and AI analyzes the accumulated data to detect unusual patterns and issue alerts.

In this process, Wicked Storm defines the role of AI as “finding and alerting.” When AI finds an anomaly signal and issues an alert, administrators and instructors review the conditions that triggered the alert together with the raw learning data, then decide how to refine the curriculum and lessons.

With this patent, both of Wicked Storm’s AI patents have been granted. The “LLM-based learner profiling method to analyze learner tendencies” (No. 10-2767110), filed on the same day, was granted in February this year. Wicked Storm earned GS Grade 1 Certification (KTL-GS-24-0035) for Lecognizer and KOLAS AI model certification (2024-122-VSW-K) for the Learning Analytics Diagnosis and Recommendation Model v1.0, and in May it received INNOBIZ (technology-innovative SME) certification (AA).

Building on its experience in public education projects, such as the Ministry of Education’s AI Digital Textbook learning data hub and KERIS’s ‘Toc Toc! Math Expedition’, and on its two patented technologies, Wicked Storm plans to broaden the foundation of its learning data analytics technology.

Adoption and demo inquiries: 02-2205-1470 · manager@wickedstorm.kr · Contact us online

Back to the newsroom