Transaction Prediction Using Noise Removal and Irregular Periodicity Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for predicting upcoming transactions are ineffective due to noise in transaction descriptions and irregularities in transaction periodicity, leading to overdrafts and inefficiencies in financial account management.
Innovation Solution
A system that utilizes a transaction prediction compute device to analyze financial transaction data, remove noise from transaction descriptions, and apply scoring algorithms that account for irregularities in transaction periodicity to accurately predict recurring transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing prediction systems are used to analyze transaction data, then the system can process financial transactions, but the detection accuracy of recurring transactions is low due to noise in descriptions and irregular periodicity
Solution Approach 1:
The patent extracts and removes noise elements from transaction descriptions to isolate the core recurring transaction patterns. By filtering out irrelevant information (noise) from transaction data, the system can more accurately identify true recurring transactions despite variations in description format and periodicity irregularities.
Solution Approach 2:
The patent transforms transaction analysis by changing parameters from exact matching of transaction descriptions to pattern-based identification. It shifts from requiring precise periodic intervals to accepting irregular periodicity within defined ranges, and from literal description matching to semantic similarity assessment, thereby improving detection accuracy.
2Reliability
If customers maintain excess funds in the financial account to avoid overdrafts, then the likelihood of overdrafts is reduced, but the efficiency of fund allocation decreases as money could be used in higher interest accounts
Solution Approach 1:
The patent performs preliminary identification and prediction of recurring transactions before they occur. By detecting patterns in advance and calculating expected transaction dates and amounts, the system enables customers to maintain only the necessary buffer balance rather than excessive funds, thereby preventing overdrafts while optimizing fund allocation efficiency.
3Productivity
If traditional prediction methods are applied to transactions with irregular periodicity, then the system can process the data, but the predictions become ineffective
Solution Approach 1:
The patent introduces dynamic parameters for detecting irregular periodicity in recurring transactions. Instead of assuming fixed intervals, the system adapts to varying transaction frequencies by calculating time windows and standard deviations, allowing accurate prediction of recurring transactions even when their periodicity changes over time.
Data Source
AI summary
Technologies for predicting recurring transactions include a compute device. The compute device includes circuitry configured to obtain data indicative of financial transactions associated with one or more customer accounts with a financial institution. The circuitry may also be configured to perform at least one data transformation operation on descriptions of the financial transactions in the obtained data to enable identification of recurring transactions, including grouping the financial transactions based on a similarity of the descriptions. Additionally, the circuitry may be configured to combine determinations from multiple analysis algorithms executed on a group of the financial transactions to obtain a confidence score indicative of whether the group of transactions represents a recurring transaction, and present, in response to a determination that the confidence score indicates that the group of financial transactions represents a recurring transaction, a predicted amount and date for a recurrence of the recurring transaction.