Handwritten Document Quick Actions via Entity Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional electronic handwriting technologies are time-consuming when it comes to gathering information or taking action about terms in handwritten documents, and they fail to provide the advantages associated with the digitization of handwriting.
Innovation Solution
A system and method that provides selectable quick actions on handwritten data by fetching and recognizing handwritten strokes, identifying keywords, classifying them into entity types, determining actionable items, and rendering these actions on the document based on user profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional electronic handwriting technologies are used to view handwritten documents, then portability is improved, but information gathering efficiency deteriorates
Solution Approach 1:
The system enables the handwritten document itself to perform actions by automatically recognizing handwritten text, classifying entities, and generating actionable items without requiring users to manually interact with external applications. The document becomes self-aware and self-acting through integrated AI processing.
Solution Approach 2:
The system integrates multiple functions including handwriting recognition, entity classification, action determination, and result rendering within a single document processing framework, allowing the document to serve multiple purposes from viewing to information gathering and action execution.
2Ease of operation
If users manually search for information in handwritten documents using conventional methods, then document viewing is possible, but operation complexity increases
Solution Approach 1:
The system merges the document viewing function with information search and action execution capabilities. Instead of requiring separate operations for viewing, searching, and acting, the system combines these functions into an integrated workflow where the document itself provides actionable information.
Solution Approach 2:
The system introduces an intermediary AI processing layer between the handwritten document and the user that automatically recognizes text, classifies entities, and determines actions, eliminating the need for users to manually navigate through multiple applications and search steps.
3Measurement precision
If handwriting recognition and entity classification are performed on all handwritten data, then information accuracy is improved, but processing time increases
Solution Approach 1:
The system applies handwriting recognition and entity classification selectively to specific regions or elements of the handwritten document that contain actionable information, rather than processing the entire document uniformly. This allows high accuracy where needed while maintaining overall processing efficiency.
Solution Approach 2:
The system performs partial processing by focusing on key entities and actionable items within the document rather than analyzing every element in detail, achieving sufficient accuracy for the most important information while reducing overall processing time.
Data Source
AI summary
A method for providing selectable quick actions on handwritten data in a handwritten document is disclosed. The method includes fetching the handwritten data corresponding to strokes made by a user in the handwritten document and recognizing text associated with the received handwritten data by employing a handwriting recognition technique. The method also includes identifying keywords in the recognized text and classifying each identified keyword into an entity type by employing a named entity recognition, a named entity linking, and/or a knowledge graph. In some embodiments, the method includes determining actionable items associated with the classified entity type corresponding to each keyword and creating quick actions, selectable by the user, for each keyword based on the determined actionable items via the knowledge graph. The method includes rendering the created quick actions associated with each keyword on the handwritten document based on a profile associated with the user.