Decoupled Computational Logic Processing with Dynamic Subgraphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computational logics in data processing applications are tightly coupled to application modules, requiring complex coding and leading to low efficiency and maintenance challenges, especially in fields like insurance where combinatory logics are intricate and frequently changing.
Innovation Solution
Decouple computational logics from input data by using a server-side computational engine that parses configuration data to generate static subgraphs, connects input data to form dynamic subgraphs, and executes them to obtain results, allowing for independent implementation, reuse, and management of computational logics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If computational logics are implemented by coding in application modules, then the computational logic can be executed, but the coupling between computational logics and application modules increases complexity and reduces maintainability
Solution Approach 1:
The patent extracts computational logics from application modules and places them in a separate computational engine. The configuration data containing computational logics is parsed independently from application code, allowing the computational engine to process multiple applications without being tightly coupled to any single application's implementation details.
Solution Approach 2:
The system segments the computational logic into distinct configuration data elements that can be independently parsed and executed. The computational engine separates the parsing of configuration data from the execution of computational logics, allowing each component to be developed, tested, and maintained independently.
2Productivity
If computational logics are coded in application modules, then the logic can be executed, but development efficiency decreases due to tight coupling and redundant coding
Solution Approach 1:
The computational engine is designed to be universal and can process computational logics from different applications through a unified configuration data format. Instead of having separate coding implementations for each application, the same computational engine parses and executes various computational logics (decision trees, arithmetic operations, aggregations) from different domains using the same framework.
Solution Approach 2:
The patent uses configuration data as a copyable representation of computational logics that can be reused across multiple applications. Once computational logic is defined in configuration data for one application, it can be copied and adapted for other applications without rewriting the underlying logic, reducing redundant coding efforts.
3Reliability
If computational logics are implemented through coding, then the computation can be performed, but maintenance becomes difficult due to frequent changes in combinatory logics
Solution Approach 1:
The system implements dynamic configuration data that can be modified without changing the application code. Computational logics are defined in a flexible configuration format that allows easy updates to decision trees, arithmetic operations, and aggregation rules. The computational engine dynamically parses these configuration changes and executes the updated logic without requiring code recompilation or redeployment.
4Ease of manufacture
If computational logics are tightly coupled to application modules, then the implementation is straightforward, but reuse of computational logics across different applications becomes difficult
Solution Approach 1:
The computational engine provides a universal interface for processing computational logics from any application. The configuration data format and parsing mechanism are designed to be application-agnostic, allowing the same computational engine to handle insurance calculations, financial computations, or other domains without application-specific customization.
Data Source
AI summary
This application relates to the field of computer software, and discloses an apparatus, method and readable storage medium for computational logic processing. The method is applied to apparatus including a server side and a client side. In this method, the client side obtains and sends input data and configuration data to the server side. The server side parses the configuration data to generate static computational subgraphs, and then connects the input data and the static computational subgraphs to generate dynamic computational subgraphs. Finally, the dynamic computational subgraphs is executed to obtain computational results. Therefore, decoupling between the computational logics and the input data is implemented, reusing of the computational logics can be implemented, and development efficiency of software developers is improved. In addition, an error of the computational logics in implementation is irrelevant with the application module because of computational logics' independent implementation, deployment, management, and operation, improving maintainability of the application module.