Automated Entity Identification for Event Probability Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional event probability prediction systems face limitations in resource efficiency, requiring extensive computer resources and interfaces with external databases, which complicates development and processing capacity, and struggle to identify anomalous behavior across multiple dimensions.
Innovation Solution
The implementation of a computer-based system using Automated Entity Identification (AEI) and Concise Profiles, which maintains only a small dynamic table in memory, replacing large disk-resident databases, and employs a recycling algorithm to identify and report entities with anomalous behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional event probability prediction systems store and access profiles for every existing entity in an external disk-resident database, then comprehensive historical information is maintained, but computer resource efficiency deteriorates and system complexity increases
Solution Approach 1:
The patent extracts only the most relevant and recently updated entity profiles from the comprehensive database into a focused in-memory data structure. This selective extraction maintains the necessary historical information for prediction while eliminating the need for complex external database interfaces, directly resolving the contradiction between comprehensive information maintenance and system complexity reduction.
Solution Approach 2:
The patent segments the entity profile storage into two parts: a comprehensive external database for long-term historical data and an in-memory data structure for actively used profiles. This segmentation allows the system to maintain comprehensive information externally while using only necessary portions internally, reducing complexity without sacrificing reliability.
2Reliability
If traditional systems maintain profiles for all entities in external databases, then complete entity data is available, but processing capacity and execution speed deteriorate
Solution Approach 1:
The patent performs preliminary action by pre-loading relevant entity profiles into the in-memory data structure before they are needed for prediction. This advance preparation ensures complete entity data is available when needed while avoiding the performance penalty of real-time database access during processing, thus maintaining data availability while improving processing capacity.
Solution Approach 2:
The in-memory data structure serves as an intermediary between the external comprehensive database and the prediction engine. This intermediary caches necessary entity profiles, allowing the system to maintain complete entity data availability through the external database while providing fast access through the in-memory cache, thereby preserving both data completeness and processing capacity.
3Ease of operation
If interfaces are created between mathematical models and external databases during development and production, then data access is enabled, but implementation complexity and development time increase
Solution Approach 1:
The patent merges the data access functionality directly into the mathematical model by incorporating an in-memory data structure within the model itself. This eliminates the need for separate external database interfaces during both development and production, enabling data access while significantly reducing implementation complexity and development time.
4Quantity of substance
If large disk-resident profile databases are used, then comprehensive entity information is stored, but resource efficiency and execution speed deteriorate
Solution Approach 1:
The patent extracts only the essential entity profiles needed for current predictions from the large disk-resident database and stores them in an in-memory data structure. This extraction maintains comprehensive entity information storage capability through the external database while dramatically improving resource efficiency by using memory instead of disk for active data, reducing I/O operations and energy consumption.
Data Source
AI summary
A computer-implemented method and system for automated entity identification for efficient profiling in an event probability prediction system. A first subset of entities belonging to one or more entity classes is defined. At least one historical profile is constructed for each entity in the subset of entities based on a set of possible outcomes of transaction behavior of each entity in the first subset of entities. Based on the historical profiles, a second subset of entities having transaction behavior associated with a transaction is selected, the transaction behavior being predictive of at least one targeted outcome from the set of possible outcomes. The first subset of entities is redefined with the second subset of entities.