Visual Query Graphs for Data Pattern Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data analysts face challenges in formulating queries that represent complex patterns in large data object collections, as specifying relationships between data objects is cumbersome and error-prone, often requiring fluency in query languages that users may not possess.
Innovation Solution
A data analysis system enables users to create visual queries using graphical user interfaces, transforming these visual representations into query templates executable by query engines, allowing users to represent queries as graphs with nodes, edges, and properties to specify data object patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If analysts formulate queries using traditional query languages to specify complex relationships between data objects, then the ability to retrieve precise data patterns is improved, but the complexity and error-proneness of the querying process increases
Solution Approach 1:
The patent introduces a visual query formulation interface that acts as an intermediary between the user and the query engine. Instead of requiring users to write complex query language code, the system provides a graphical interface where users can visually construct queries by selecting data objects and specifying relationships through point-and-click interactions. This intermediary layer translates the visual specifications into executable query language, thereby maintaining query accuracy while dramatically reducing formulation complexity
Solution Approach 2:
The patent replaces the mechanical syntax-based query formulation process with a visual, graphical interface. Instead of requiring users to manually type and structure query language syntax, the system uses graphical elements such as icons, buttons, and visual selectors to represent data objects and relationships. This substitution eliminates syntax errors and reduces the cognitive load associated with learning and remembering query language syntax, while preserving the ability to express complex data relationships
2Ease of operation
If analysts use visual query interfaces to simplify query formulation, then ease of operation is improved, but the ability to express complex query patterns may be limited
Solution Approach 1:
The patent implements a dynamic visual query interface that adapts its capabilities based on the data schema and user selections. As users build queries by selecting data objects, the interface dynamically presents relevant relationship options, filters, and parameters specific to those objects. This dynamic behavior allows the interface to maintain simplicity for basic queries while automatically providing access to advanced query patterns when needed, thus preserving both ease of operation and adaptability
Solution Approach 2:
The patent segments the query formulation process into distinct, manageable visual components. Each data object type is represented by separate visual elements, and relationships between objects are established through discrete graphical actions. This segmentation allows users to build complex queries incrementally by combining simple visual building blocks, making the interface both easy to use and capable of expressing sophisticated query patterns through composition
Data Source
AI summary
A method and apparatus for a data analysis system for analyzing data object collections is provided. The data analysis system includes one or more graphical user interfaces comprising various interface elements that enable users to create visual queries. A visual query is constructed as a graph representing a pattern of interest in a collection of data objects. A visual query may include one or more graph elements and property information associated with the specified graph elements. After a user has constructed a visual query, the system may transform the visual query into a query template. A query engine may then execute the query template to search a data object collection for data object results corresponding to the specified pattern. The search for instances of a specified pattern in a collection of data objects is referred herein to as a “search around.”