AI Transaction Attribute Extraction for User-Readable Payment Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems are unable to read and understand transaction data, such as ACH transaction data, which is heavily abbreviated and concatenated, leading to user confusion and a high volume of transaction inquiries and disputes, burdening payment service computing platforms.
Innovation Solution
Utilizing a trained AI model to process transaction data in a computer-readable format, determining transaction attributes, and presenting them in a user-readable format, thereby reducing user confusion and disputes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If transaction data is kept in computer-readable format (abbreviated and concatenated), then data processing efficiency is improved, but user understanding of transactions deteriorates
Solution Approach 1:
The patent introduces an intermediary system (AI model or translation service) that translates between computer-readable abbreviated formats and user-readable explanations. This intermediary layer allows the system to maintain efficient compressed data storage while providing clear user-friendly transaction descriptions, resolving the contradiction between processing efficiency and user understanding.
Solution Approach 2:
The system changes the representation parameter of transaction data by maintaining the original abbreviated computer-readable format for processing while generating alternative user-readable representations. This parameter transformation allows the same data to serve both efficient processing and clear user communication purposes simultaneously.
2Device complexity
If conventional systems process transaction data, then system simplicity is maintained, but user confusion and disputes increase
Solution Approach 1:
The system enables self-service by automatically translating and explaining transaction data without requiring user intervention to understand the abbreviated formats. The AI model or translation service autonomously processes the data and provides clear explanations, reducing user confusion and disputes while maintaining relative system simplicity.
Solution Approach 2:
The system incorporates feedback mechanisms where user interactions with transaction information can be used to refine and improve the translation and explanation models over time, reducing misunderstandings and disputes while keeping the core system architecture simple.
3Loss of information
If transaction data is translated to user-readable format, then user understanding is improved, but processing time increases
Solution Approach 1:
The system performs preliminary translation and explanation generation when transaction data is first received or accessed, so that when users need to view the information, the user-readable format is already prepared. This preliminary action reduces the time required for users to understand their transactions without significantly impacting overall processing efficiency.
Solution Approach 2:
The system applies partial translation only when and where needed (e.g., only for transaction descriptions that require clarification, or only when users request detailed explanations), rather than translating all data continuously. This selective approach provides user understanding when needed while minimizing the time cost of translation processing.
4Measurement precision
If AI models are used to process transaction data, then transaction attribute determination accuracy is improved, but computational resources consumed increase
Solution Approach 1:
The system uses AI models partially by applying them only to transactions that require enhanced analysis or when high accuracy is critical, rather than processing all transactions through the full AI pipeline. This selective application maintains high measurement precision for important transactions while reducing overall computational resource consumption.
Solution Approach 2:
The system changes the computational parameter by using lighter-weight processing methods for routine transactions and reserving heavy AI model computation only when necessary for complex or high-value transactions. This parameter adjustment optimizes the balance between determination accuracy and computational resource consumption.
Data Source
AI summary
Processing transaction data using artificial intelligence (AI) is described. A payment service computing platform may receive transaction data associated with users of a payment application, wherein the transaction data is received in a computer-readable format, and the payment service computing platform may provide a prompt to a trained AI model, wherein the prompt includes a portion of the transaction data that represents a transaction associated with a user of the users. The payment service computing platform may determine, based at least in part on the trained AI model processing the prompt, one or more attributes of the transaction, and cause information indicative of the one or more attributes to be presented via the payment application executing on a user device of the user, wherein the information is presented (i) in a graphical user interface and (ii) in a user-readable format instead of the computer-readable format.