Query Rewrites for Auto-Complete Suggestion Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines are inefficient in providing query completion suggestions, especially for users entering long search queries or those with query terms not found in popular searches, leading to increased processing time before relevant results can be obtained.
Innovation Solution
A method and system that includes a server system with processors and memory to receive partial search queries, search a database of complete queries, and provide autocomplete suggestions by rewriting the query if initial matches do not meet predefined criteria, ensuring timely and relevant results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system searches a database of complete search queries to identify matching queries, then the accuracy of query completion suggestions is improved, but the processing time increases
Solution Approach 1:
The system performs preliminary actions by maintaining a database of complete search queries that have been pre-processed and indexed. When a user enters a partial query, the system can quickly search this pre-prepared database for matching complete queries, rather than generating suggestions from scratch in real-time. This preliminary preparation of the query database enables fast retrieval and accurate matching without incurring high processing delays during actual use.
Solution Approach 2:
The system creates copies of complete search queries from historical data and stores them in a database for rapid retrieval. Instead of analyzing and generating query suggestions in real-time, the system copies and retrieves pre-existing complete queries that match the user's partial input. This copying approach allows the system to provide accurate completions by matching against stored copies of actual user queries, significantly reducing processing time while maintaining high accuracy.
2Ease of operation
If the system provides autocomplete suggestions for long search queries with many terms, then the user experience is improved, but the complexity of the matching algorithm increases
Solution Approach 1:
The system segments the complete search queries into individual terms or phrases, allowing it to handle long queries with multiple terms more effectively. By breaking down complex queries into smaller components, the system can match user input against individual terms in the database without requiring the entire long query to match exactly. This segmentation simplifies the matching algorithm while still providing accurate completions for multi-term queries, thereby improving user experience without proportionally increasing algorithmic complexity.
Solution Approach 2:
The system applies partial matching rather than requiring complete query matching. When a user enters a partial search query, the system matches only the entered portion against the database of complete queries, ignoring whether the complete query contains additional terms. This partial action approach allows the system to provide useful completions even for long queries without requiring complex algorithms to analyze and match every term, thus maintaining ease of operation while controlling algorithmic complexity.
3Reliability
If the system rewrites partial search queries to produce rewritten queries, then the relevance of suggestions is improved, but the processing complexity increases
Solution Approach 1:
The system changes parameters of the partial search query by rewriting it in different forms or with alternative term combinations. Instead of simply matching the exact user input, the system rewrites the query with modified parameters such as synonym substitution, term reordering, or grammatical variations. This parameter changing approach enables the system to find relevant complete queries that might not match the original input exactly, thereby improving the relevance and reliability of suggestions while using controlled rewriting rules to manage processing complexity.
Data Source
AI summary
A computer-implemented method for proving complete query suggestions, performed by a server system, includes receiving a partial search query from a client system. A database of complete search queries is searched to identify a first set of matching complete search queries. If the first set of matching complete search queries fails to satisfy predefined criteria (e.g., with respect to the quantity and/or quality of the matching complete search queries), a rewrite of the received partial search query is performed to produce a first set of rewritten partial search queries. The database of complete search queries is searched to identify a second set of complete search queries matching any of the first set of rewritten partial search queries. At least a subset of the second set of matching complete queries are sent to the client system.