Search System Attribute Hierarchy for Query Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional search systems fail to effectively present relevant attributes of items in search results, leading to user frustration and increased computing resource consumption due to repetitive queries and listings, as users often need to sift through numerous results to find desired attributes.
Innovation Solution
A search system that determines and ranks attributes using a machine learning model trained on user data and search history, providing a hierarchy of attributes and visual indications within search results to facilitate quick identification of relevant items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional search systems present only basic search results without attribute hierarchy, then the search results interface is simple, but users experience frustration and require repetitive queries to find desired attributes
Solution Approach 1:
The system performs preliminary analysis of item attributes and user search intent before presenting search results. A machine learning model pre-ranks attributes and pre-identifies relevant attribute values, so that when search results are displayed, the most relevant attributes are already highlighted and organized, eliminating the need for users to perform repetitive filtering and querying actions.
Solution Approach 2:
The patent introduces an intermediary machine learning model that acts as a mediator between the search query and the search results. This model analyzes both the search intent and item attributes, then generates a ranked hierarchy of attributes that bridges the gap between user needs and item information, presenting only the most relevant attributes to users.
2Loss of information
If conventional search systems require users to sift through numerous results to find desired attributes, then all item attributes are available, but computing resource consumption increases due to repetitive queries
Solution Approach 1:
The system extracts only the most relevant attributes from the complete set of item attributes based on user search intent. The machine learning model identifies and extracts the top-ranked attributes that are most likely to help users make purchasing decisions, presenting only these extracted attributes in the search results rather than requiring users to filter through all available attributes.
Solution Approach 2:
The patent applies local quality by providing different attribute presentations for different items based on their specific characteristics and relevance to the search query. Each item's search result displays a customized set of attributes ranked by relevance, rather than a uniform presentation of all attributes, optimizing the information display for each specific context.
3Loss of information
If conventional search systems display all attributes equally, then no attribute information is lost, but users cannot quickly identify relevant items
Solution Approach 1:
The system segments the complete set of item attributes into a ranked hierarchy based on relevance to user search intent. Attributes are divided into tiers or levels of importance, with the most relevant attributes displayed prominently in the search results. This segmentation allows users to quickly identify relevant items based on the top-ranked attributes while knowing that additional attribute information is available if needed.
Data Source
AI summary
A search system determines a hierarchy of attributes for a set of search results and returns the search results based on a top-ranked attribute. The search system identifies item listings based on a search input and determines attributes of the item listings. A machine learning model generates a hierarchy of the attributes. The search results are provided based on a top-ranked attribute from the hierarchy of the attributes.