Query Suggestion System Using Completeness Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines fail to provide effective query suggestions that are tailored to the completeness of a search string, often offering either irrelevant specific suggestions for short or non-specific strings and lacking utility for users who need assistance with longer queries.
Innovation Solution
A system that determines a measure of query completeness by analyzing user input and provides either general or specific query suggestions based on a threshold, using a query log to calculate probabilities and entropies to offer relevant suggestions, such as geographic locations or category labels, before the user submits their query.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If specific query suggestions are provided for all search strings, then users with complete queries receive irrelevant suggestions, but users with incomplete queries lack useful guidance
Solution Approach 1:
The system dynamically adjusts the type of query suggestions provided based on the measured completeness of the search string. When query completeness is high, specific suggestions are offered; when low, general suggestions are provided. This dynamic adaptation resolves the contradiction by making the suggestion system both relevant and universally applicable across different query states.
Solution Approach 2:
The system changes the parameter of suggestion specificity based on the query completeness metric. By thresholding on the completeness measure, the system transitions between providing general category suggestions and specific query completions, thereby achieving both relevance for complete queries and applicability for incomplete ones.
2Adaptability or versatility
If general query suggestions are provided for all search strings, then users with incomplete queries receive useful guidance, but users with specific queries receive less precise help
Solution Approach 1:
The system dynamically adjusts suggestion specificity based on real-time assessment of query completeness. This dynamic behavior allows the system to maintain broad applicability through general suggestions while preserving precision when users have already entered specific query terms.
Solution Approach 2:
The system modifies the specificity parameter of suggestions based on the completeness metric of the input query. When completeness exceeds a threshold, specific suggestions are provided; otherwise, general suggestions are offered, balancing applicability and precision.
3Ease of operation
If query suggestions are provided without assessing completeness, then the system is simple to implement, but the suggestions are not tailored to user needs
Solution Approach 1:
The system performs preliminary assessment of query completeness before generating suggestions. This advance evaluation allows the system to select the appropriate suggestion type, improving user experience without requiring complex real-time adjustments during suggestion delivery.
Solution Approach 2:
The query completeness metric serves as an intermediary that bridges the simple suggestion generation mechanism and the nuanced user needs. This intermediate measurement enables tailored suggestions while maintaining relative system simplicity.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing query suggestions. In one aspect, a method includes receiving a search string from a user device. A measure of query completeness is determined for the search string. The measure of query completeness is compared to a threshold measure of query completeness. One or more specific query suggestions are provided to the user device, selected from a plurality of specific query suggestions for the search string, when the measure of query completeness exceeds the threshold measure of query completeness; or one or more general query suggestions are provided to the user device, selected from a plurality of general query suggestions for the search string, when the measure of query completeness does not exceed the threshold measure of query completeness.