Shelf-Specific Facet Extraction for Accurate Query-to-Shelf Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques fail to determine relevant facets related to search queries, as these queries do not directly correspond to browse pages, leading to inefficient item filtering on websites.
Innovation Solution
A system and method for shelf-specific facet extraction that includes pre-processing to generate a facet universe, characterizing keywords, and performing fuzzy matching to identify relevant facets for browse shelves, enabling real-time extraction of facets specific to the search query.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional techniques are used to handle search queries, then the system maintains simplicity in query processing, but the system fails to determine relevant facets related to search queries
Solution Approach 1:
The patent introduces browse shelves as intermediary structures that connect search queries to relevant facets. The system identifies one or more browse shelves corresponding to a search query, then extracts facets from those shelves. This intermediary approach enables the system to determine facet relevance without requiring direct complex analysis between queries and facets, thus improving measurement precision while managing system complexity.
Solution Approach 2:
The system performs preliminary organization of facets into browse shelves before processing search queries. Facets are pre-categorized and structured in browse shelves that correspond to different product categories and attributes. When a search query is received, the system can quickly identify relevant browse shelves and extract facets without performing complex real-time analysis, thereby improving detection accuracy while maintaining processing efficiency.
2Measurement precision
If search queries are directly mapped to facets without browse shelves, then the process is simpler, but the mapping accuracy and relevance are insufficient
Solution Approach 1:
The system pre-organizes facets into browse shelves that represent different product categories and attributes. This preliminary structuring allows the system to efficiently map search queries to relevant facets by identifying corresponding browse shelves, rather than performing exhaustive analysis of all facets for each query. This approach improves mapping accuracy while maintaining extraction efficiency through the pre-established organizational structure.
Solution Approach 2:
Browse shelves serve as intermediary structures that facilitate accurate query-to-facet mapping. Instead of directly mapping queries to facets, the system identifies relevant browse shelves first, then extracts facets from those shelves. This intermediary approach improves mapping accuracy by providing a structured intermediate layer that captures the semantic relationship between queries and facets, while the systematic process maintains productivity.
3Loss of information
If the system extracts all facets for every search query, then completeness is achieved, but the information overload reduces user experience
Solution Approach 1:
The system extracts only the relevant facets from browse shelves that correspond to the search query, rather than returning all available facets. By identifying specific browse shelves related to the query and extracting facets only from those shelves, the system provides complete relevant information without overwhelming users with unrelated facet data, thus improving ease of operation while maintaining information completeness for the specific query context.
Solution Approach 2:
The system applies local quality by providing different facet sets tailored to each specific search query based on the identified browse shelves. Instead of a uniform approach for all queries, the system customizes the facet extraction to match the specific context and intent of each query, ensuring that users receive the most relevant and useful facet information for their particular search needs, thereby enhancing user experience.
Data Source
AI summary
A method including obtaining one or more predicted shelves corresponding to the keyword query. The method additionally can include generating linked categorical facets corresponding to the one or more predicted shelves based on shelf-categorical facet linkages. The method further can include generating, using fuzzy matching, candidate shelf-specific facets based on shelf-specific facet representation mappings and the linked categorical facets. The method additionally can include determining one or more shelf-specific facets from the candidate shelf-specific facets based on facet information in the candidate shelf-specific facets. The one or more shelf-specific facets can correspond to one or more shelves of the one or more predicted shelves. The method further can include outputting the one or more shelves and one or more respective shelf-specific facets of the one or more shelf-specific facets that correspond to each of the one or more shelves. Other embodiments are described.