Empirical Search Query Replacement for Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Search engines often return incorrect results due to misspelled, inaccurate, or poorly correlated search queries, leading to inefficient and time-consuming user searches, especially when information is organized differently than expected, and conventional query suggestion methods fail to adapt to database contents and search history.
Innovation Solution
A computer-implemented method that generates alternative search query terms based on empirical data from prior search sessions, evaluating replacement terms for their occurrence rates and conversion rates to suggest terms that have led to desired outcomes, such as purchases, thereby improving search efficiency by recommending terms that are likely to yield relevant results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional query suggestion methods are used that generate alternatives based on semantic similarity, then typo corrections may be provided, but the suggestions remain impertinent when the initial query directs to an unintended category and do not adapt to database contents
Solution Approach 1:
The system pre-collects and stores empirical search session data including query sequences and outcome events in a database before new searches occur. This preliminary data accumulation enables the system to have replacement term statistics ready when users submit queries, eliminating the need for real-time analysis and allowing immediate adaptation to database contents and search patterns.
Solution Approach 2:
The system uses outcome events from search sessions (such as purchase actions or leave-without-purchase actions) as feedback to evaluate replacement terms. By analyzing whether subsequent queries led to desired outcomes, the system continuously refines its understanding of effective query replacements and adapts suggestions to both database contents and actual user behavior patterns.
2Reliability
If users manually formulate multiple different search queries to find desired information, then they may eventually locate relevant results, but the process becomes non-economical, time-consuming, and frustrating
Solution Approach 1:
The system automatically generates and provides replacement query suggestions without requiring users to manually iterate through multiple search attempts. By analyzing empirical data and outcome events, the system self-services users by predicting effective query replacements and presenting them immediately, eliminating the need for users to spend time formulating and testing multiple queries themselves.
Solution Approach 2:
The system performs preliminary analysis of search patterns and outcome events to pre-identify effective replacement terms before users need them. This advance preparation allows the system to immediately present high-probability replacement queries when users submit initial queries, significantly reducing the time users would otherwise spend on iterative searching.
3Productivity
If information is organized and indexed in databases in a manner that deviates from common-sense expectation, then the database structure may be optimized for storage or retrieval, but average users find it difficult to formulate representative query expressions
Solution Approach 1:
The system acts as an intermediary between users and the database by translating user-friendly initial queries into effective database search terms. Instead of requiring users to understand database organization and indexing structures, the system analyzes empirical data to identify replacement terms that bridge the gap between common-sense user expressions and database-optimized query structures, making the system easy to operate while maintaining high retrieval efficiency.
4Measurement precision
If the system automatically replaces user input terms with replacement terms, then more relevant search results are yielded, but the system must accurately evaluate and select from multiple candidate terms
Solution Approach 1:
The system evaluates candidate replacement terms using quantifiable parameters derived from empirical data, such as occurrence rates and outcome event rates. By transforming the complex task of query term evaluation into measurable statistical parameters, the system can objectively compare and select the most relevant replacement terms without requiring complex qualitative analysis, thereby achieving high precision while managing system complexity through parameterization.
Data Source
AI summary
Systems and methods for automatically generating replacement query terms that offer improved search efficiency. Recommended search query terms are generated based on statistic information derived from empirical data recording prior search sessions with respect to searching on a search engine. A query term entered later in a search session is treated as a possible replacement term for a query term entered earlier in the same session. Upon receiving an initial query term in a new search session, the replacement terms of the initial query term are identified from the empirical data and evaluated as candidates for replacing the initial query term in the new search session. The evaluation is based on the respective occurrence rates that the candidates are used as replacement terms for the initial query term in the empirical data, and based on the respective conversion rates of the candidates.