AI Document Preparation Support with Retrieval and Check Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The efficiency of document preparation is low due to the reliance on manual work and visual confirmation, especially in documents requiring high accuracy, and existing automatic generation methods fail to meet accuracy requirements.
Innovation Solution
A document preparation support device that includes a retrieval unit for related documents, a generation unit for tentative documents, and an extraction unit for check items, utilizing generative AI to assist in document preparation and review.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual work and visual confirmation are used for document preparation, then document accuracy is maintained, but preparation time and effort increase significantly
Solution Approach 1:
The patent introduces an AI assistant as an intermediary between the author and the document approval process. The AI generates tentative documents, extracts check items, and creates correction proposals, serving as a mediator that handles routine tasks while human reviewers focus on accuracy-critical decisions. This resolves the contradiction by automating time-consuming tasks without compromising document quality.
Solution Approach 2:
The document preparation process is segmented into distinct tasks: AI generates tentative documents, extracts check items, and creates correction proposals, while human reviewers handle approval and final validation. This segmentation allows automated handling of routine operations and human oversight of critical accuracy requirements, simultaneously improving efficiency and maintaining reliability.
2Productivity
If language model is used for automatic document generation, then preparation time is reduced, but document accuracy cannot be satisfied sufficiently
Solution Approach 1:
The patent implements a feedback mechanism where the AI extracts check items from the tentative document and generates correction proposals based on reviewer feedback. This closed-loop system allows the AI to learn from errors and improve document accuracy over time while maintaining high preparation efficiency. The feedback loop ensures that accuracy issues are systematically addressed without sacrificing productivity.
3Reliability
If approval process is added to language model-generated documents, then document accuracy is improved, but time and effort are still consumed
Solution Approach 1:
The AI performs preliminary actions by generating tentative documents, extracting check items, and creating correction proposals before human approval. This preliminary processing reduces the workload and time required for the approval process, as reviewers only need to validate and approve rather than create documents from scratch. The preliminary action maintains accuracy while minimizing time loss.
Data Source
AI summary
A document preparation support device includes a retrieval unit that retrieves a related document corresponding to text included in a document preparation command input by an author, a generation unit that generates a tentative document on the basis of the document preparation command and the retrieved related document, and an extraction unit that, on the basis of the content of a draft document generated from the tentative document by the author, extracts a check item to be reviewed by a document reviewer in the draft document.