Asynchronous Unstructured Data Processing for Real-Time Decisioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional automated decisioning systems for unstructured data face inefficiencies due to sequential processing, leading to delays, timeouts, and reduced flexibility, especially when handling exceptions and varying data formats, which hinder real-time decisioning capabilities.
Innovation Solution
An automated adjudication architecture that separates the data conversion and decisioning processes into asynchronous components, allowing for modular, parallel execution through a conversion component that extracts structured data segments and a decisioning component that processes these segments independently, with interface components managing data transfer without interrupting the workflow.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If sequential two-step processing is used for unstructured data, then processing simplicity is maintained, but processing speed and real-time decisioning capability deteriorate
Solution Approach 1:
The patent segments the sequential two-step processing into multiple parallel processing streams, where data extraction and decisioning operations are divided into independent tasks that can execute concurrently. This segmentation enables the system to process unstructured data through multiple pathways simultaneously, dramatically improving processing speed while maintaining manageable complexity through modular task design.
Solution Approach 2:
The patent transitions from sequential one-dimensional processing to parallel multi-dimensional processing by introducing temporal and spatial dimensions to the processing architecture. Multiple extraction and decisioning components operate in parallel across different processing lanes, transforming the single-file sequential workflow into a multi-dimensional concurrent processing system that achieves real-time decisioning.
2Adaptability or versatility
If data extraction and decisioning are tightly coupled, then data flow control is simplified, but system flexibility and adaptability deteriorate
Solution Approach 1:
The patent segments the tightly coupled extraction-decisioning system into loosely coupled modular components. Each extraction component and decisioning component is separated into independent units with well-defined interfaces, allowing individual components to be modified, updated, or replaced without affecting the entire system. This modular segmentation enables flexible adaptation to different data formats and decisioning requirements.
Solution Approach 2:
The patent introduces intermediary mechanisms such as message queues and event buffers between extraction and decisioning components. These intermediaries decouple the direct dependency between components, allowing extraction to proceed independently and decisioning to consume data at its own pace. This intermediary layer provides the flexibility needed to handle varying data formats and exception scenarios while maintaining system coherence.
3Measurement precision
If processing delays occur at data extraction, then data accuracy may be improved through thorough extraction, but decisioning component idle time increases
Solution Approach 1:
The patent implements preliminary action by having multiple decisioning components prepared and ready to process data before extraction is complete. While one extraction component is processing complex unstructured data to ensure accuracy, other decisioning components remain in standby or process previously extracted data, eliminating idle time and ensuring continuous operation without compromising extraction thoroughness.
Solution Approach 2:
The patent maintains continuity of useful action through parallel processing streams and load balancing mechanisms. When extraction delays occur in one stream, other streams continue processing, and decisioning components dynamically shift workload to maintain continuous operation. This ensures that no decisioning component remains idle while extraction accuracy is preserved through dedicated extraction resources.
4Adaptability or versatility
If processing is limited to particular data formats, then processing efficiency is improved, but system versatility and exception handling capability deteriorate
Solution Approach 1:
The patent implements universality by designing extraction and decisioning components with multi-functional capabilities that can handle multiple data formats and exception scenarios. The system incorporates universal data validation and normalization layers that adapt to different input formats while maintaining consistent processing efficiency. This universal design enables the system to process diverse unstructured data including images, PDFs, and text documents without sacrificing performance.
Solution Approach 2:
The patent utilizes parameter changes by dynamically adjusting processing parameters based on detected data formats and characteristics. When unusual data formats or exception scenarios are detected, the system automatically modifies extraction and decisioning parameters to optimize handling of the specific data type. This parameter adaptation maintains processing efficiency across varying data formats while enhancing system versatility.
Data Source
AI summary
Various embodiments of the present disclosure provide an automated asynchronous decisioning process that improves the functionality of a computer in various aspects. The process may comprise receiving, an input data object that comprising a set of unstructured data elements and converting the input data object to a structured data object by (i) extracting, using an optical character recognition model, a set of input features from the input data object, (ii) generating, using a structuring ruleset, a structured data element based on the set of input features, and (iii) storing the structured data element within a polling queue and in association with the request identifier. The process may comprise receiving, from a polling component, a polling request and responsive to receiving the polling request, providing, to the polling component, at least one of the request identifier, the structured data element, or a conversion status for the structured data object.