AI Tutor Knock Adds Red-Ink Correction for Handwritten Math Answers
The feature moves AI tutoring from text-based feedback to a visual correction format that mimics traditional teacher grading, aiming to improve learning motivation and reduce teacher workload.
Reporting from 1 sources: ASCII.jp.
Hanji Co. has added a red-ink correction feature to its AI tutor Knock. Students photograph handwritten answers and receive visual red-ink markings on the image, identifying errors and providing advice within tens of seconds. The feature supports up to university entrance exam level. It was developed by analyzing hundreds of student answers and consulting teachers to replicate the familiar teacher correction experience.
Knock previously offered text-based correction, but student interviews showed that screen readability and comfort directly affect learning motivation. The new red-ink feature lets students upload a photo of their handwritten answer and receive visual markings on the image, identifying errors and providing advice within tens of seconds. It supports advanced descriptive answers at the university entrance exam level. The development team reviewed hundreds of actual high school student answers and held discussions with teachers to build a database of common stumbling points and teacher-specific grading patterns. Hanji also plans to expand the feature to track class-level learning trends from correction data.
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.
Sources
- ASCII.jp AIチューター「Knock」に数学の赤入れ添削機能を追加