Behavior Model Segmentation for Accurate Entity Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data engines face challenges in accurately generating scores for entities, leading to unreliable measures and classifications, which can result in inappropriate service adjustments.
Innovation Solution
A system that includes a data engine with modules for data collection and processing, behavior modeling, score generation, and classification, which generates behavior models and scores based on past and future events to predict entity performance and adjust services accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional data engines generate scores based on predetermined period events, then the scoring process is simple and fast, but the accuracy and reliability of the scores deteriorate
Solution Approach 1:
The patent segments the scoring process into multiple independent behavior models (first behavior model for first time span, second behavior model for second time span, third behavior model for third time span). Each model processes specific time-span events separately and generates individual scores, which are then combined to produce the final score. This segmentation improves score accuracy by capturing different temporal patterns while maintaining manageable complexity through modular model structures.
Solution Approach 2:
The patent introduces a temporal dimension by creating behavior models for multiple distinct time spans (first, second, and third time spans) rather than using a single predetermined period. This multi-dimensional temporal approach allows the system to analyze entity behavior across different time horizons, improving score accuracy by capturing both short-term and long-term patterns while adding structural depth to the data engine.
2Reliability
If data engines use multiple time spans for behavior modeling, then score accuracy improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies preliminary action by pre-defining multiple behavior models for different time spans before actual scoring occurs. The first behavior model processes first time span events, the second behavior model processes second time span events, and the third behavior model processes third time span events. These models are prepared in advance with their respective scoring logic, so when scoring is needed, the system can efficiently retrieve and apply the appropriate pre-configured models, reducing real-time processing overhead while maintaining high reliability through comprehensive multi-time-span analysis.
3Measurement precision
If the system generates multiple behavior models for different time spans, then prediction accuracy improves, but the device complexity increases
Solution Approach 1:
The patent segments the behavior modeling function into distinct first, second, and third behavior models, each responsible for a specific time span. This segmentation allows each model to be specialized for its time span requirements, improving prediction accuracy by capturing time-specific behavioral patterns. The modular segmented structure manages complexity by organizing models into clear, independent units with defined responsibilities rather than requiring a single monolithic complex model.
Solution Approach 2:
The patent creates behavior models that serve multiple functions: the first behavior model handles first time span events, the second behavior model handles second time span events, and the third behavior model handles third time span events. Each model is multi-functional in that it can process various types of events within its designated time span while contributing to the overall score. This universality approach improves prediction accuracy through comprehensive coverage while managing complexity by using consistent model frameworks across different time spans.
Data Source
AI summary
Various systems, mediums, and methods for providing services may involve data engines configured to generate scores associated with one or more entities and then to classify the entities based on the scores. The data engine may collect data and based on the collected data may generate a first behavior model in a first time span, and a second and third behavior models in a second time span. The data engine may generate a first score based on the first behavior model, a second score based on the second behavior model, and a third score based on the third behavior model. The data engine may generate a final score based on the first, second, and third scores. The data engine can classify the entity based on the final score. The data engine can then automatically adjust one of the services provided to the entity based on the final score.