AI Request Enrichment for Real-Time Ledger Exception Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data management systems require tedious manual effort to sift through high volumes of operational data for identifying operational exceptions, insights, and anomalies, and lack real-time interaction with financial data via natural language prompts.
Innovation Solution
A data management system utilizing structured data return templates and large language models to automatically identify and respond to user requests for application functionality, enabling real-time detection of anomalies and insights from transactional data, and performing operations like journal entry creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual methods are used to sift through high volumes of operational data, then data analysis can be performed, but the process is tedious and time-consuming
Solution Approach 1:
The patent replaces manual mechanical data sifting with an AI-powered system that uses large language models to automatically analyze operational data, identify anomalies, and generate insights. The system processes data through automated prompt engineering and LLM inference, eliminating the need for manual review while maintaining analytical accuracy.
Solution Approach 2:
The system enables self-service data analysis by allowing users to interact with operational data through natural language prompts. The AI system automatically executes complex data queries, identifies relevant information, and presents insights without requiring users to manually navigate through vast datasets, thus freeing user time while maintaining analytical capability.
2Ease of operation
If real-time interaction with financial data via natural language is enabled, then operational efficiency improves, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary layer consisting of prompt templates and structured data return templates that mediate between user natural language queries and the underlying complex data systems. This intermediary translates simple user prompts into structured LLM requests and formats responses according to predefined schemas, simplifying the interaction interface while managing system complexity through abstraction.
Solution Approach 2:
The system implements a universal prompt template framework that can handle multiple types of data queries and operational requests through a single interface. The same prompt engineering infrastructure supports various data management tasks including anomaly detection, insights generation, and journal entry creation, reducing the need for separate specialized systems for each function.
3Extent of automation
If automated AI processing is implemented, then manual effort is eliminated, but initial system setup and configuration complexity increases
Solution Approach 1:
The patent employs preliminary action through pre-configured prompt templates and structured data return templates that are prepared in advance. These templates define the expected input formats, processing logic, and output structures before actual data processing occurs. This preliminary setup enables automated processing to proceed smoothly during operation without requiring complex real-time configuration, as the automation infrastructure is pre-assembled with all necessary components.
Data Source
AI summary
Systems, methods, and computer-readable media are provided for accessing a user request received in a user session with an application. Item(s) of data may be selected in the user session in association with the user request. A data management system determines that the request is for or otherwise relevant to item(s) of application functionality such as financial exceptions, account balances, and/or operations detail. The data management system generates a prompt by adding structured data return template(s) for triggering the item(s) of application functionality, the user request, and, if applicable, structured text representing the item(s) of data to a prompt template. The data management system prompts a large language model with the prompt and receives a result. The result includes a data structure conforming to the structured data return template(s) and that is based at least in part on the item(s) of data. The data management system triggers the relevant item(s) of application functionality based at least in part on the result and causes display of information indicating the item(s) of application functionality have been triggered.