Relationship Visualization for Nonhomogeneous Entities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Document management systems often generate overwhelming search results that do not effectively guide users to the desired information, especially when searching for specific details like names or phone numbers without remembering the associated context, leading to the need to open multiple documents to find the relevant information.
Innovation Solution
A visualization system that uses a multipartite graph to illustrate relationships between users and documents, selecting and displaying contextual information from entity nodes based on significance scores and date ranges to provide a clear, concise representation of how the user is related to the document or other entities, such as people, projects, or topics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a pattern-matching search mechanism is used to find documents containing a desired phrase, then the search can locate documents mentioning the phrase, but the user cannot find information when the search phrase is not remembered (such as a person's name or telephone number)
Solution Approach 1:
The patent introduces an intermediary entity (such as a person, project, or topic) that connects the user to the desired information. Instead of directly searching for the forgotten detail, the user searches for the intermediary entity, and the system retrieves associated information including the forgotten detail. For example, searching for a person's name retrieves their telephone number, or searching for a project retrieves related documents and entities.
2Quantity of substance
If traditional search results are generated displaying file attributes and titles, then the search mechanism can locate matching documents, but the results overwhelm the user and do not help identify which document contains the desired information
Solution Approach 1:
The patent extracts and displays only the most relevant and distinctive information from search results, rather than showing all available attributes. It identifies and highlights key entities and relationships that directly help the user identify the target document, removing extraneous information that contributes to overwhelming the user.
Solution Approach 2:
The patent applies different display qualities to different parts of the search results based on their relevance. Important entities and relationships are highlighted or emphasized, while less important information is downplayed or omitted, creating a non-uniform display that guides user attention to the most useful information.
3Device complexity
If the visualization displays only file attributes like location and type, then the search results can be sorted and organized, but valuable contextual information that helps the user make a decision is not displayed
Solution Approach 1:
The patent nests multiple levels of information within the visualization structure. File attributes are nested within entity nodes, which are nested within relationship paths, which are nested within the overall search result visualization. This hierarchical nesting allows contextual information to be displayed in an organized manner without overwhelming the user.
Data Source
AI summary
A relationship visualization system displays contextual information for a relationship between two entities of a document management system, such as for a user and a document. The system can receive a request for a relationship visualization from a user, such that the request indicates the user and a document. Then, the system determines, from a multipartite graph, a set of relationship paths coupling entity nodes that correspond to the user and the document. A relationship path can include one or more entity nodes indicating contextual information for the relationship between the user and the document. Then, the system selects a first group of entity nodes from the set of relationship paths to represent the relationship between the user and the document. The system then provides a relationship visualization that displays contextual information from the first group of entity nodes to illustrate the relationship between the user and the document.