Graph-Based Information Search Linking Structured and Unstructured Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to utilize unstructured data associated with structured data, such as graph representations and experimental condition information, for analysis and data utilization.
Innovation Solution
An information search device that includes a graph drawing unit to draw structured data, a graph extraction range acquisition unit to extract data points, and an in-range graph extraction data acquisition unit to acquire a table with links to unstructured data, along with a search expression acquisition unit and data conversion units to facilitate the integration and analysis of structured and unstructured data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If only structured data is utilized for analysis, then data processing is simple and straightforward, but unstructured data including graph representations and experimental condition information cannot be utilized
Solution Approach 1:
The patent segments data into structured data (stored in databases) and unstructured data (stored separately but linked through graph representations). This segmentation allows the system to handle different data types appropriately while maintaining the ability to access both structured and unstructured data through the graph interface, thereby improving data utilization capability without overwhelming system complexity.
Solution Approach 2:
The patent introduces graph representations as an intermediary between structured data and unstructured data. The graph serves as a mediator that links experimental data points with their corresponding unstructured data (such as experimental conditions and graph images), enabling the system to utilize both data types effectively while managing complexity through a unified access mechanism.
2Ease of operation
If experimental data is stored with link information to experiment notebook data, then data retrieval is improved, but there is no mechanism for users to analyze experimental data and identify data to be utilized
Solution Approach 1:
The patent adds a visual dimension to data retrieval by implementing graph representations that display experimental data points graphically. This dimensional change from purely tabular data to visual graph representation enables users to easily retrieve data while simultaneously providing analytical capabilities, as users can visually identify trends, patterns, and specific data points of interest directly from the graph interface.
Solution Approach 2:
The graph representation serves as an intermediary interface between the stored experimental data and the user. It provides both retrieval functionality (through searchable data points) and analysis capability (through visual representation), bridging the gap between simple data storage and complex data analysis requirements.
3Device complexity
If all experimental data is stored in a unified database, then data storage is simplified, but unstructured data cannot be effectively integrated and accessed
Solution Approach 1:
The patent segments the storage system into multiple components: structured data storage (for numerical experimental data), unstructured data storage (for experimental conditions and graph images), and graph representation storage. This segmentation allows each data type to be stored in the most appropriate format while maintaining simplified access through unified graph-based retrieval mechanisms.
Solution Approach 2:
The patent implements a nested storage structure where graph representations contain references to both structured data points and unstructured data elements. The graph acts as a container that nests multiple data types together, allowing unified access to heterogeneous data while maintaining the organizational benefits of segmented storage.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information search device, method, and program for acquiring a link to unstructured data associated with structured data are provided. The information search device includes a graph drawing unit configured to draw as structured data, a graph extraction range acquisition unit configured to extract one or more data points in the drawn graph, and an in-range graph extraction data acquisition unit configured to acquire a table including structured data including the extracted data point and a link to unstructured data.