Terminal Screen Word Extraction for Search Automation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Workers face difficulties in extracting relevant information from terminal screens due to the lack of registered words in search engines, leading to hindered work operations, as existing methods fail to identify related information when unknown words are searched.

Innovation Solution

A data processing apparatus and method that acquires words from terminal screen content and determines their relevance using an inverted index database, allowing only registered words to be searched, thereby ensuring accurate information retrieval.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If a search engine is introduced to help workers find information, then information accessibility is improved, but workers still cannot find information unless they know the exact search keywords

Engineering Contradiction:
Improveinformation accessibilityVSAvoidsearch operation ease
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The system automatically extracts keywords from the terminal screen content and performs searches without requiring worker intervention for keyword selection. The search engine serves itself by autonomously identifying search terms from the displayed content, eliminating the need for workers to manually input keywords.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system introduces an intermediary component that sits between the terminal screen and the search engine. This intermediary automatically extracts meaningful keywords from the screen content and translates them into effective search queries, bridging the gap between the worker's information need and the search engine's requirements.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If workers repeatedly search for information using different keywords, then they may eventually find relevant information, but work efficiency is reduced due to repeated searching

Engineering Contradiction:
Improveinformation retrieval completenessVSAvoidwork efficiency
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The system performs preliminary keyword extraction from the terminal screen content before the search operation. By pre-identifying relevant keywords automatically, the system eliminates the need for workers to engage in repeated trial-and-error searching, thus improving work efficiency while ensuring comprehensive information retrieval.

Inventive Principle:
Principle #10Preliminary action

3Extent of automation

If existing methods extract words from terminal screen content for searching, then search automation is improved, but information cannot be extracted when words are not registered in the search engine

Engineering Contradiction:
Improvesearch automationVSAvoidinformation extraction accuracy
Core Design Contradiction:
Extent of automationVSLoss of information

Solution Approach 1:

The system incorporates a feedback mechanism that checks whether extracted keywords yield relevant search results. If no relevant information is found, the system adjusts its keyword extraction strategy by selecting alternative words from the terminal screen content, ensuring that information extraction accuracy is maintained even when initial keyword choices are not registered in the search engine.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS12099530B2Search using words from user interface
Publication Date: 2024.09.24 NIPPON TELEGRAPH & TELEPHONE CORP
  • US12099530B2 patent drawing
  • US12099530B2 patent drawing
  • US12099530B2 patent drawing

AI summary

A data processing apparatus includes an acquisition unit that acquires words included in a displayed content of a terminal screen, and a determination unit that determines that the words acquired by the acquisition unit are search candidates in a case where the acquired words are included in an inverted index in which words and identification information of work data including the words are stored in association with each other.