Session-Based Query Suggestions Using Similarity Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines provide query suggestions based solely on popularity, which often ignore the context of previous queries within a user's search session, leading to irrelevant suggestions.
Innovation Solution
A system that compares current search queries with previous sessions to identify similar query paths and suggests queries that are frequently used in similar sessions, providing context-based suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If query suggestions are based solely on popularity of search queries, then the system is simple to implement, but the relevance of suggestions to user context deteriorates
Solution Approach 1:
The patent segments the search session into individual queries and identifies similar sessions by matching sequences of queries. This allows the system to provide context-aware suggestions without requiring complex overall session analysis, balancing simplicity with relevance.
Solution Approach 2:
The system pre-processes and stores search session data in a query log, organizing it by session and query sequence. This preliminary organization enables efficient retrieval and comparison of similar sessions during live operation without complex real-time processing.
2Reliability
If query suggestions consider previous queries in the current session and similar previous sessions, then the relevance of suggestions improves, but the system complexity increases
Solution Approach 1:
The patent creates simplified representations of search sessions by storing sequences of queries in a query log. Instead of analyzing complete session contexts, the system works with copied query sequences, enabling efficient similarity comparison through straightforward matching operations.
3Measurement precision
If the system stores and analyzes complete search session data, then the accuracy of similarity identification improves, but the data processing requirements increase
Solution Approach 1:
The patent extracts only the essential elements for similarity comparison - the sequences of search queries - from complete search sessions. By taking out and storing only these query sequences in the query log, the system maintains accuracy for identifying similar sessions while significantly reducing data processing requirements compared to analyzing complete session data.
Data Source
AI summary
Methods, systems, and apparatus, including computer program products, in which one or more search query suggestions are made for a current search session. Similar previous search sessions which include search queries common to the current search session are identified. Based upon the similar previous search sessions, one or more suggested search queries are derived and provided to a search engine interface for serving to a user or a client.