Task-Specific Document Presentation via Keyword Extraction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing business document presentation systems struggle to effectively present relevant documents to operators handling varying tasks, as they often extract universal information like 'customer name' and 'address,' making it difficult for operators to find task-specific documents.

Innovation Solution

A business-document presenting device that extracts and compares text information from the operator's screen with stored data to identify unique keywords, using techniques like IDF and crawling, to search and present task-specific documents without requiring keyword settings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If business documents including frequently appearing words from operation screens are extracted, then universal information such as customer name and address can be presented, but task-specific documents become difficult to find due to lack of differentiation

Engineering Contradiction:

Improvedocument relevanceVSAvoidtask variation coverage

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies local quality by differentiating document extraction based on task-specific characteristics. Instead of uniformly extracting all frequent words, the system identifies and extracts words that are specific to particular tasks (local characteristics) versus universal words (global characteristics). This allows the system to present documents tailored to each specific task context, improving both relevance and adaptability simultaneously.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent segments the document extraction process into two distinct components: extraction of universal information (common across all tasks) and extraction of task-specific information (unique to each task). By dividing the keyword extraction into these segments, the system can present a combination of both universal and task-specific documents, resolving the contradiction between presenting reliable universal documents and adapting to varied tasks.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If keyword search is required to find business documents, then search precision can be improved, but operator workload increases due to manual keyword setting

Engineering Contradiction:

Improvesearch accuracyVSAvoidoperator convenience

Core Design Contradiction:

Measurement precisionVSEase of operation

Solution Approach 1:

The patent implements self-service by enabling the system to automatically extract task-specific keywords and search for relevant business documents without requiring manual keyword input from operators. The system autonomously analyzes the operation screen, identifies task-specific words, and performs the search, thereby maintaining high search accuracy while eliminating the burden of manual keyword setting.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent applies preliminary action by pre-extracting and storing task-specific keywords from operation screens before the search is needed. The system proactively identifies and caches relevant keywords associated with different tasks, so when a search is required, the system can immediately use these pre-prepared keywords without requiring real-time manual input, thus improving both accuracy and ease of operation.

Inventive Principle:

Principle #10Preliminary action

3Quantity of substance

If all words in text information are compared for document extraction, then comprehensive coverage is achieved, but processing time increases significantly

Engineering Contradiction:

Improveinformation coverageVSAvoidprocessing duration

Core Design Contradiction:

Quantity of substanceVSLoss of time

Solution Approach 1:

The patent applies the extraction principle by selectively removing and focusing only on the most relevant portion of text information - specifically, task-specific keywords - rather than processing all words. The system extracts only the essential keywords that differentiate tasks from the full text information, thereby maintaining comprehensive coverage of task-specific content while dramatically reducing the volume of data that requires processing.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements partial action by processing only a subset of text information (task-specific keywords) rather than all words. Instead of performing exhaustive comparison of every word in the text, the system performs partial processing on the extracted keywords, achieving sufficient information coverage for accurate document retrieval while minimizing processing time through selective action.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentEP3955194B1Business document presenting device, business document presenting method, and business document presenting program

Publication Date: 2024.04.17 NIPPON TELEGRAPH & TELEPHONE CORP

Figure 1
Figure 2
Figure 3

AI summary

A storage unit (14) stores text information displayed on a terminal screen during the processing of tasks in the past. An acquisition unit (15a) acquires text information displayed on the terminal screen. An extraction unit (15b) compares the acquired text information and the text information stored in the storage unit (14) and extracts different items of text information among tasks as keywords for searching for business documents.