Predictive Model Database with Segmented Access Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data analysis software lacks an efficient method to build and utilize predictive models for estimating future events based on user input, particularly in decomposing data streams and filtering noise, while also allowing for model optimization and accessibility in a searchable database.
Innovation Solution
A method is developed to decompose data input streams, build predictive models using precursor data, and store them in a searchable database, utilizing techniques like Bayesian algorithms, Markov rules, and genetic algorithms for pattern recognition, and allowing users to query models for future event probabilities through a search engine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If predictive models are built and stored in a searchable database for user access, then model accessibility and reuse are improved, but system complexity and data management overhead increase
Solution Approach 1:
The system segments the predictive modeling functionality into distinct components: model building modules, model storage database, model search interface, and model execution engines. Each component operates independently but connects through standardized interfaces, allowing the system to scale and manage complexity through modular architecture.
Solution Approach 2:
The patent introduces intermediary layers including a model repository that mediates between model creators and model users, and a search engine that acts as an intermediary between user queries and stored models. These intermediaries simplify the interaction complexity while enabling broad accessibility.
2Measurement precision
If data streams are decomposed and noise filtering is applied to build accurate predictive models, then model accuracy is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary data stream decomposition and noise filtering during the model building phase rather than during model execution. Data preprocessing, feature selection, and initial model training are completed in advance, allowing the deployed models to execute quickly when making predictions without repeating the intensive processing steps.
Solution Approach 2:
The patent applies selective noise filtering and data decomposition only to the extent necessary for achieving acceptable model accuracy. The system identifies and processes only the critical data elements that significantly impact prediction quality, rather than exhaustively processing all data, thus balancing accuracy with processing efficiency.
Data Source
AI summary
Techniques to estimate the probability of a future event occurring are described. The techniques include decomposing a data input stream to build a database of precursor data and building predictive models using the precursor data. Also disclosed are techniques in which by using a search engine to search a database of models to find a model and a user can query a found model to develop an inference of the likelihood of the future event.