Intelligent Forecasting System Data Aggregation and Enrichment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprises face challenges in efficiently aggregating and normalizing vast amounts of data from disparate sources due to incompatible database structures, requiring extensive resources and expertise for data translation and analysis, and often resort to costly custom software development for forecasting tasks.
Innovation Solution
A system and method that allows users to select statistical forecasting models, train time series models, and generate predictions through a user interface framework without writing custom software, automating model selection, parameter tuning, and outlier detection, enabling data aggregation and enrichment across various datasets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional brute force methodologies are used to create custom software for translating different database types, then data compatibility is improved, but resource requirements and expertise needs increase significantly
Solution Approach 1:
The patent introduces an intermediary translation layer that automatically converts data between different database types using predefined mapping rules. This intermediary component handles the complexity of data translation without requiring custom software development for each database type, thus maintaining data compatibility while reducing overall system complexity
Solution Approach 2:
The system changes the approach from creating custom software (structural parameter) to configuring translation rules (parameter-based configuration). By using parameterized translation rules that can be adjusted without rewriting software, the system achieves adaptability across different database types while keeping the core software architecture simple and reusable
2Ease of manufacture
If common data models are used to create uniform data structure, then data organization is improved, but customization requirements and expertise needs increase
Solution Approach 1:
The patent implements a dynamic common data model that can be configured and customized through parameters rather than requiring structural changes. The model adapts to different enterprise needs by allowing dynamic adjustment of data fields, relationships, and constraints without requiring expert intervention to modify the underlying model structure
Solution Approach 2:
The data model is segmented into standardized core components and customizable extension components. The core common data model provides uniform organization, while separate configurable modules allow customization for specific enterprise requirements, reducing the need for expert-level model modification
3Adaptability or versatility
If custom software is created for forecasting tasks, then forecasting functionality is improved, but development time and costs increase
Solution Approach 1:
The patent creates a universal forecasting platform that can handle multiple forecasting tasks across different data types and business scenarios. This single multi-functional system replaces the need for separate custom software developments for each forecasting requirement, significantly reducing development time while maintaining adaptable forecasting capabilities through configurable parameters
Data Source
AI summary
An intelligent forecasting system that includes sources of financial and non-financial input data and a user interface generator for generating user interfaces having window elements that display the input data. The window element includes a persistent navigation pane element having vertically stacked actuatable navigation soft buttons for accessing one or more portions of an intelligent forecasting. The navigation soft buttons include, among other buttons, an Outlier Treatment soft button for processing the input data by applying thereto a statistical processing model to detect outliers in the input data, a Model Selection soft button for selecting a statistical forecasting model to apply to the input data, a Signal Explorer soft button for selecting a signal transformation method for transforming the input data, a Model Prediction soft button for selecting one of the statistical forecasting models to apply to the input data, and a Simple Prediction soft button for automatically generating forecasts.