Product Attribute Normalization via Classification Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional normalization rules applied to source data in online retail systems often result in inconsistent and incorrect values, leading to different sizes for un-normalized values expressed differently, despite being equal in dimension.
Innovation Solution
A system and method that utilize a domain-specific language to extract structured values from raw source data, optimize normalization rules, and integrate these into a runtime normalization process, including the use of a multi-class classification algorithm to train a classification model for determining additional normalization rules and validating them, ensuring consistent and accurate normalization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional normalization rules are applied to source data, then data processing can be performed, but the normalization results are inconsistent and incorrect
Solution Approach 1:
The patent transforms raw source values into structured values through domain-specific language parsing, then applies normalization rules to produce normalized values. This multi-stage parameter transformation resolves the contradiction by ensuring consistent interpretation of different value representations (e.g., different unit formats) while maintaining processing capability.
Solution Approach 2:
The patent introduces structured values as an intermediary between raw source values and normalized values. The domain-specific language parser creates this intermediate structured representation that standardizes the data format before normalization, ensuring consistent results regardless of the original value representation.
2Adaptability or versatility
If multiple normalization rules are used to handle different value formats, then more data types can be processed, but the number of rules increases complexity
Solution Approach 1:
The patent segments the normalization process into distinct stages: parsing raw values into structured values using domain-specific language, then applying normalization rules to structured values. This segmentation allows comprehensive coverage of different data formats through the parser while keeping normalization rules simpler and more manageable.
Solution Approach 2:
The domain-specific language parser serves as a universal interface that handles multiple value formats (different units, representations) by converting them all into a standardized structured format. This multi-functional parser reduces the need for multiple specialized normalization rules, simplifying the overall system.
3Ease of manufacture
If manual creation and maintenance of normalization rules is performed, then rules can be customized, but the process is time-consuming and not scalable
Solution Approach 1:
The system enables self-service through automated parsing of domain-specific language and machine learning-based classification that automatically determines normalization rules. The machine learning model can be retrained with new data to adapt to changing requirements without manual rule creation, reducing maintenance time while preserving customization capability.
Solution Approach 2:
The patent incorporates machine learning classification that uses feedback from normalized data to automatically refine and update normalization rules. This feedback mechanism allows the system to learn from actual usage patterns and improve its normalization accuracy over time without manual intervention, reducing maintenance time.
Data Source
AI summary
Systems and methods including one or more processing modules and one or more non-transitory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of receiving a plurality of normalization rules, using the plurality of normalization rules and a multi-class classification algorithm to train a classification model for one or more product attributes of a plurality of products, determining additional normalization rules for transforming additional structured values to additional final normalized values based on a classification of un-normalized additional raw source values with the classification model, validating the additional normalization rules, and adding the additional normalization rules to the classification model if the additional normalization rules are validated.