Smart Time Series Analytics Software Model Development
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current software platforms for time series and machine learning model development are complex, requiring extensive coding and statistical knowledge, and lack features for assumption testing, enhanced exploration, and business input integration, leading to inefficiencies and increased model risk due to overfitting, data quality issues, and lack of transparency.
Innovation Solution
The Smart Time Series Analytics Software (STSA) and Machine Learning Way (MLWay) streamline model development by automating data validation, feature creation, model selection, and reporting, providing customizable configurations and enhanced exploration capabilities, allowing non-technical users to build and validate robust models with improved interpretability and governance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional time series and machine learning model development processes are used, then model development can be performed with basic software tools, but the process becomes complex requiring extensive coding and statistical knowledge
Solution Approach 1:
The patent introduces an intermediary software platform that sits between the user and the complex modeling processes. This platform provides automated model development, assumption testing, and validation capabilities, shielding users from the underlying complexity while maintaining access to sophisticated analytical methods. The intermediary handles data preprocessing, model selection, and evaluation automatically, reducing the need for users to have extensive coding and statistical knowledge.
2Productivity
If automated model fitting is implemented, then model development speed increases, but assumption testing and enhanced exploration capabilities are lost
Solution Approach 1:
The patent implements preliminary action by automatically performing assumption testing, data quality checks, and model validation steps before the main model fitting process. This ensures that models are built on sound assumptions and high-quality data from the outset, improving robustness without slowing down the overall development process. The system pre-processes data, tests assumptions, and validates models in an automated sequence that maintains both speed and reliability.
3Manufacturing precision
If extensive exploration and testing are performed to ensure model quality, then model precision improves, but development time increases from 4-12 months
Solution Approach 1:
The patent enables self-service model development where the system automatically performs exploration, testing, validation, and optimization without requiring extensive manual intervention. The automated platform conducts comprehensive model quality checks, sensitivity analyses, and performance evaluations autonomously, maintaining high precision while dramatically reducing development time from months to days or hours. The system serves itself by automatically iterating through model configurations and selecting the best performing models.
4Adaptability or versatility
If manual model development with coding is used, then customization flexibility is available, but the process lacks transparency and governance capabilities
Solution Approach 1:
The patent implements comprehensive feedback mechanisms that automatically track, record, and report all model development decisions, assumptions, and results. The system provides transparent documentation of the entire modeling process, including data quality metrics, assumption test results, model performance evaluations, and validation outcomes. This feedback loop ensures that customization flexibility is maintained while simultaneously improving transparency and governance through automated tracking and reporting capabilities.
Data Source
AI summary
A process implemented as software for building, developing, and enhancing a model for use in forecasting, having a first user input step, wherein a user to input data using a user interface on a user device and providing the user input data to an application program interface (API); the API performs an auto data validation step, a feature creation step comprising using domain knowledge to extract features from raw training data; a feature encoding step comprising using the created features and raw training data to train different candidate models; a model selection step wherein the user is prompted to select a best model from the number of trained candidate models based on user defined model rankings; a best model review step comprising producing detailed information on the best model through statistical diagnostics, sensitivity, back-test and performance analysis; and generating implementation code for the best model; processing a set of data to be analyzed using the best model, forecasting an outcome based on processing the set of data to be analyzed with the best model, and providing the forecast to a user by a user interface on a user device.

