Structure Learning Parser for Purchase Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The diversity of payment confirmation formats and merchants makes it difficult for individuals to track their purchases and for merchants to obtain comprehensive customer purchase history data, limiting targeted marketing efforts.
Innovation Solution
A system and method for extracting product purchase information from electronic messages using a structure learning parser that automatically learns message formats, allowing for precise extraction and aggregation of purchase data across various formats and languages, enabling enhanced purchase history tracking and targeted marketing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple different payment confirmation formats from diverse merchants are used, then more comprehensive purchase tracking information is obtained, but the difficulty and cost of extracting product purchase information increases
Solution Approach 1:
The patent applies universality by creating a single extraction system that can handle multiple different confirmation formats from diverse merchants. The system uses a unified approach with regular expressions and data normalization techniques to process various message types (email, SMS, push notifications) and formats (different merchants, different layouts) through one multi-functional platform, eliminating the need for separate extraction systems for each merchant or format type.
2Measurement precision
If a structure learning parser is used to automatically learn message formats, then extraction precision is improved, but the initial setup time and data requirements increase
Solution Approach 1:
The patent applies preliminary action by collecting and storing sample confirmation messages from various merchants before deployment. The system pre-processes these samples to learn common patterns, formats, and structures. This preliminary learning phase enables the extraction system to achieve high precision from the start without requiring extensive setup time during actual operation, as the structure learning has already been performed on the training data.
3Loss of information
If purchase confirmations from diverse merchants are aggregated, then comprehensive customer profiles can be developed, but the difficulty of tracking and organizing information increases
Solution Approach 1:
The patent applies homogeneity by normalizing extracted purchase information into a unified data structure. All purchase confirmations from diverse merchants are transformed into a consistent format with standardized fields (product name, price, date, merchant, quantity). This homogenization of data structures makes the aggregated information easy to track, organize, and analyze, regardless of the original diversity of confirmation formats.
Data Source
AI summary
Improved systems and methods for extracting product purchase information from electronic messages transmitted between physical network nodes to convey product purchase information to designated recipients. These examples provide a product purchase information extraction service that is able to extract product purchase information from electronic messages with high precision across a wide variety of electronic message formats and thereby solve the practical problems that have arisen as a result of the proliferation of different electronic message formats used by individual merchants and across different merchants and different languages. In this regard, these examples are able to automatically learn the structures and semantics of different message formats, which accelerates the ability to support new message sources, new markets, and different languages.