Automatic Data Model Generation for Disparate Data Sources
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Organizations face challenges in managing and analyzing large volumes of disparate data sources due to unfamiliar arrangements, requiring advanced skills and knowledge to develop data models, leading to duplication and inefficiencies in data design.
Innovation Solution
A system utilizing modeling engines to automatically generate data models and visualizations based on search expressions, recommending data fields and visualizations through user interfaces, enabling users to create data models without deep technical knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If analysts manually create data models from disparate data sources, then data models can be customized to specific needs, but the process requires advanced data design skills and deep knowledge of underlying data systems
Solution Approach 1:
The patent introduces an intermediary system (the platform with automated data model generation capabilities) that mediates between the analyst and the complex data sources. This intermediary handles the complexity of data discovery, mapping, and model generation automatically, allowing analysts to create data models without needing deep technical knowledge of the underlying data systems while still achieving customized results.
2Adaptability or versatility
If analysts manually develop data models, then data models can be tailored to specific analytical needs, but duplication of data designs occurs due to difficulties in discovering previously provided data models
Solution Approach 1:
The patent implements preliminary action by automatically discovering and indexing previously created data models and data sources before the analyst needs them. The system proactively searches through the organization's data landscape, identifies relevant existing models, and presents them to the analyst, eliminating the need for analysts to manually search and recreate similar models from scratch.
Solution Approach 2:
The system provides feedback by displaying previously created data models and data sources to the analyst during the data model creation process. This feedback mechanism allows analysts to see what has already been done, avoid duplication, and build upon existing work, thereby saving time and reducing redundant efforts.
3Quantity of substance
If organizations collect and organize large volumes of data from disparate sources, then comprehensive data analysis becomes possible, but the data becomes unfamiliar and irrelevant to analysts without advanced skills
Solution Approach 1:
The patent applies self-service by enabling the system to automatically discover, map, and generate data models from the organization's data sources without requiring manual intervention for each step. The system serves itself by autonomously navigating the data landscape, identifying relationships, and creating usable data models, thereby making the large volumes of data accessible and relevant to analysts without requiring advanced technical skills.
Data Source
AI summary
A method for recommending data visualizations includes selecting candidate data fields based on a value of a data field meeting threshold criteria. A first user input is received, selecting one or more of the candidate data fields. One or more recommended types of data visualizations are generated and displayed based on the selected candidate data fields. A second user input is received, selecting a type of visualization from the one or more recommended types of data visualizations. A data visualization that includes at least one of the one or more candidate data fields associated with the selected type of visualization is generated and displayed based on the selected type of visualization.