Rule Engine Segmentation for Massive Personalized Rule Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current open-source rule engines are limited in processing and loading large numbers of rules, leading to slower speeds and memory overruns, making them unsuitable for massive personalized rule applications, especially in scenarios like e-commerce where numerous rules need to be applied to large-scale groups.
Innovation Solution
The method involves classifying rules into general operators (general laws) to support massive customized actions for entities, improving initialization speed and allowing for dynamic hot loading, updating, and deleting of rules, thereby enabling efficient customization of personalized rules for massive entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional rule engines are used to process massive personalized rules, then rule processing capability is provided, but processing speed decreases and memory overruns occur
Solution Approach 1:
The patent segments the massive rule set into two distinct layers: general rules that apply to all entities and personalized rules that are specific to individual entities. This segmentation allows the system to process only relevant rules for each entity rather than loading and processing all rules, thereby maintaining high processing speed while supporting massive numbers of personalized rules.
Solution Approach 2:
The patent extracts and separates personalized parameters from the general rule structure. By taking out the personalized components and storing them independently, the system avoids loading unnecessary personalized data into memory during general rule processing, preventing memory overruns while still enabling massive personalized rule customization.
2Quantity of substance
If traditional rule engines are used to process massive personalized rules, then rule processing capability is provided, but memory capacity is exceeded
Solution Approach 1:
The patent segments rule storage into multiple components: general rules stored in a rule engine and personalized rules stored separately in databases or configuration files. This segmentation allows the system to load only essential general rules into memory while keeping personalized rules available on-demand, thus supporting massive rule quantities without exceeding memory capacity.
Solution Approach 2:
The patent extracts personalized parameters and rules from the core rule engine memory space and stores them in external storage. This extraction enables the system to reference massive numbers of personalized rules without loading them all into memory simultaneously, preventing memory overruns while maintaining full rule availability.
3Adaptability or versatility
If massive personalized rules are customized for massive entities, then customization capability is achieved, but system complexity increases
Solution Approach 1:
The patent creates a universal rule structure where general rules can apply to multiple entities across different domains, and personalized parameters can be added or removed based on specific needs. This universality allows the system to handle diverse customization requirements without proportionally increasing system complexity, as the same framework serves multiple purposes.
Solution Approach 2:
The patent introduces an intermediary layer that manages the mapping between general rules and personalized parameters. This intermediary abstraction simplifies the complexity by providing a standardized interface for rule customization, allowing users to configure personalized rules without directly managing the underlying complex data structures and processing logic.
Data Source
AI summary
The present application discloses a method, apparatus, device and storage medium for customizing personalized rules for entities, and relates to the fields of big data and cloud storage. The specific implementation scheme is: setting corresponding multiple fit conditions for entities which need to be fitted with personalized rules; where each fit condition includes a general law, a personalized parameter, and a corresponding customized action; loading data of multiple first entities corresponding to the multiple fit conditions; and for each first entity, acquiring a corresponding fit condition of the first entity; and extracting, from the data of the first entity, the personalized parameter in the corresponding fit condition, and determining a corresponding customized action according to the extracted personalized parameter and the general law in the corresponding fit condition.