Chat-Guided Search Interface for Short Query Intent Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional keyword-based search methods fail to capture user intent and semantic relationships, leading to poor search performance in high-dimensional embedding spaces, especially when user inputs are short and lack contextual information.
Innovation Solution
Utilizing large language models (LLMs) to enhance user inputs by generating complementary data, such as additional keywords or synonyms, and transforming the input into vector embeddings for improved search relevance, incorporating user-specific context and preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional keyword-based search methods are used, then the search system is simple and fast, but the search accuracy and ability to capture user intent deteriorate
Solution Approach 1:
The patent introduces large language models as an intermediary component between the user's short query and the search system. The LLM enriches the query by generating additional keywords, synonyms, and contextual information, transforming simple keyword matching into a more sophisticated semantic search process that improves accuracy without requiring complete system redesign
Solution Approach 2:
The system performs preliminary processing of user queries using LLMs to generate expanded search terms and contextual data before the actual search execution. This preliminary enrichment of search queries with additional keywords and semantic information prepares more comprehensive search inputs, improving result relevance while keeping the core search mechanism relatively simple
2Measurement precision
If user inputs are short and lack contextual information, then the user interaction is simple and fast, but the search relevance deteriorates
Solution Approach 1:
The LLM acts as an information bridge that receives short user inputs and generates expanded contextual content including related keywords, synonyms, and semantic variations. This intermediary process compensates for the lack of contextual information in brief user queries while maintaining the simplicity of user interaction
Solution Approach 2:
The system transforms the parameter density of search queries by using LLMs to generate multiple expanded terms and contextual variations from short inputs. This parameter expansion converts sparse short queries into rich multi-term search expressions, improving relevance without requiring users to provide detailed information
Data Source
AI summary
A computer-implemented method is disclosed. The method includes: receiving, via a first user interface, a selection associated with an object; determining a first set of object attributes based on the selection; presenting, via a second user interface, a text prompt for a user to identify a subset of the first set of object attributes; receiving, via the second user interface, an indication of one or more preferred object attributes of the identified subset; and updating the first user interface to display content relating to objects associated with the one or more preferred object attributes.