Product Search Classification via Intermediary Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional internet-based product search engines face issues with aggregating product information from multiple merchants, leading to erroneous or conflicting results due to varied classification of product data across different merchants.
Innovation Solution
A computer-implemented method for classifying search results by identifying key words and structural parameters, such as product type, brand, and retailer, and updating the product database based on search queries, while validating and storing textual content with associated structural parameters to provide accurate and consistent search results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If product information is aggregated from multiple merchant websites using web-crawlers, then the quantity of product data increases, but the accuracy and consistency of product information decreases due to varied classification across merchants
Solution Approach 1:
The patent introduces an intermediary classification system that mediates between diverse merchant product data and the search engine. This intermediary layer standardizes product information by mapping various merchant classifications to a unified taxonomy, resolving conflicts and inconsistencies while preserving the quantity of aggregated data.
Solution Approach 2:
The system changes the classification parameters by transforming merchant-specific product categories into standardized parameters. This parameter transformation allows diverse data sources to be aggregated while maintaining consistency, as each merchant's varied classification scheme is converted to a common parameter set.
2Adaptability or versatility
If merchants use different product descriptions and categories to establish unique website interfaces, then the adaptability of individual merchant websites improves, but the reliability of aggregated product information deteriorates
Solution Approach 1:
The patent segments the product classification system into two independent layers: the merchant-specific presentation layer that maintains unique interfaces and adaptability, and the standardized data layer that ensures reliable aggregated information. This segmentation allows each layer to fulfill its specific function without compromising the other.
Solution Approach 2:
An intermediary mapping layer is introduced that translates merchant-specific categories and descriptions into standardized classifications. This mediator preserves the adaptability of individual merchant websites while ensuring that aggregated product information maintains reliability through consistent parameter mapping.
3Ease of operation
If traditional search engines directly match search queries to product webpages using description tags, then the simplicity of the search process is maintained, but the precision of search results decreases due to erroneous or conflicting product information
Solution Approach 1:
The patent introduces an intermediary classification and validation layer between the search query and product database. This intermediary processes and validates product information before indexing, improving search result precision while maintaining the simplicity of the user search process through automated backend processing.
Solution Approach 2:
The system performs preliminary classification, validation, and standardization of product information during data aggregation, before the actual search operation. This preliminary action ensures that search results are precise without requiring complex processing during the user's search interaction, thus maintaining ease of operation.
Data Source
AI summary
A product search engine uses web-crawling software to index textual webpage content from multiple internet sources. The product data obtained from the web-crawling process is then granulized by parsing methods into key words, such as, words and/or phrases. These components are compared with existing key words from search queries or webpage content. Each key word includes component tags, where the component tags map each component to product classification information in the form of structural parameters; and other product information such as, URLs, product images and product descriptions. When at least one matching key word is found, structural parameters are extracted from the matching components and assigned to the received key words. A search results webpage including product information grouped by associated structural parameter may be provided.