17454729. SEQUENCE LABELING TASK EXTRACTION FROM INKED CONTENT simplified abstract (MICROSOFT TECHNOLOGY LICENSING, LLC)
SEQUENCE LABELING TASK EXTRACTION FROM INKED CONTENT
Organization Name
MICROSOFT TECHNOLOGY LICENSING, LLC
Inventor(s)
Apurva Sandeep Gandhi of Union City CA (US)
Gilbert Antonius of San Ramon CA (US)
Tra My Nguyen of Brighton MA (US)
Ryan Serrao of Seattle WA (US)
SEQUENCE LABELING TASK EXTRACTION FROM INKED CONTENT - A simplified explanation of the abstract
This abstract first appeared for US patent application 17454729 titled 'SEQUENCE LABELING TASK EXTRACTION FROM INKED CONTENT
Simplified Explanation
The patent application describes a computer system that can receive user input in the form of inked content on a digital canvas. The system processes this content to identify different writing regions, each containing recognized text and layout features.
- The system tokenizes a target writing region into a sequence of tokens.
- It uses a task extraction subsystem to process the tokens, considering both the recognized text and layout features.
- The target writing region is segmented into sentence segments.
- Each sentence segment is classified as either a task sentence or a non-task sentence.
Potential Applications
- Digital note-taking applications that can automatically identify and categorize tasks within handwritten notes.
- Collaborative document editing tools that can extract and organize tasks from shared documents.
- Intelligent personal assistants that can understand and act upon tasks mentioned in handwritten or digital content.
Problems Solved
- Manual identification and categorization of tasks within handwritten or digital content can be time-consuming and error-prone.
- Extracting tasks from complex documents with varying layouts and formats can be challenging for traditional text processing algorithms.
- Understanding the context and relationships between tasks and other content elements can be difficult without considering layout features.
Benefits
- Improved productivity by automating the identification and organization of tasks within handwritten or digital content.
- Enhanced collaboration and task management in shared documents by extracting and tracking tasks.
- More efficient and accurate personal assistants that can understand and act upon tasks mentioned in various forms of content.
Original Abstract Submitted
A computer system is provided that includes one or more processors configured to receive user input for inked content to a digital canvas, and process the inked content to determine one or more writing regions. Each writing region includes recognized text and one or more document layout features associated with that writing region. The one or more processors are further configured to tokenize a target writing region of the one or more writing regions into a sequence of tokens, process the sequence of tokens of the target writing region using task extraction subsystem that operates on tokens representing both the recognized text and the one or more document layout features of the target writing region, segment the target writing region into one or more sentence segments, and classify each of the one or more sentence segments as a task sentence or a non-task sentence.
