ML-Based Intermediary for Multiplatform Query Coordination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital infrastructure in large institutions, comprising disparate and siloed platforms, faces challenges in coordinating interactions to enhance accuracy and efficiency, particularly when servicing customers, leading to increased complexity and resource allocation issues as platforms evolve and interact more frequently.
Innovation Solution
An intermediary automated system powered by machine learning modules, which select and generate queries to determine properties of specific platforms, reducing the load on back-end systems and enabling more accurate and efficient customer navigation by training models to interact with event handlers and manage workflows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple disparate platforms are maintained independently to provide diverse services, then service functionality and versatility are improved, but system complexity and coordination difficulty increase
Solution Approach 1:
The patent introduces an intermediary coordination layer that sits between multiple disparate platforms and customers. This intermediary automatically routes customer requests to the appropriate platforms, manages data flow, and coordinates interactions without requiring direct integration between all platforms. The intermediary acts as a mediator that simplifies the complex web of platform interactions while preserving the functionality of each independent platform.
2Adaptability or versatility
If platforms are updated frequently to improve services, then service quality and adaptability are improved, but coordination stability and complexity management deteriorate
Solution Approach 1:
The coordination system performs preliminary actions by pre-establishing communication protocols, data schemas, and interaction patterns between platforms before updates occur. The system maintains version information and coordination rules that can accommodate platform updates without requiring simultaneous updates across all platforms. This preliminary structuring allows individual platforms to be updated independently while maintaining overall system stability.
3Ease of operation
If parallel platforms are created to improve customer navigation, then ease of operation is improved, but resource allocation efficiency and system complexity worsen
Solution Approach 1:
The patent implements a universal coordination platform that performs multiple functions: routing customer requests, managing data flow between platforms, providing navigation assistance, and coordinating updates. Rather than creating separate parallel systems for each function, this single multi-functional coordination layer handles all navigation and coordination tasks, reducing the total number of platforms needed while improving ease of operation.
4Measurement precision
If comprehensive platform coordination is implemented to improve accuracy, then query accuracy and service precision are improved, but computational resources and system complexity increase
Solution Approach 1:
The coordination system implements local quality by applying different coordination strategies and levels of scrutiny to different types of queries and platforms. Rather than uniformly coordinating all platform interactions with the same level of complexity, the system adapts its coordination approach based on the specific query type, platform characteristics, and data sensitivity. This localized approach maintains high accuracy for critical queries while reducing computational overhead for routine operations.
Data Source
AI summary
A system and method are provided for coordinating resources in multiplatform environments. The illustrative method includes providing a first platform to receive a first query for determining one or more properties of a process of a second platform of an enterprise. The method includes selecting a machine learning model from a plurality of machine learning models based on the first query, and generating a second query, based on the first query, for a selected machine learning model associated with the process. The second query is provided to the selected machine learning model. The selected machine learning model searches the second platform to determine properties, having been trained on queries from intermediate platforms. The selected machine learning model outputs one or more determined properties in response to the second query. The one or more determined properties are served as a response to the first query.