Query Completion Demotion for Search Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Formulating a query that accurately represents a user's informational need can be challenging for search engines, as users often receive unsuitable query completions during the query entry process due to the lack of dynamic refinement of suggested options.
Innovation Solution
A method that dynamically refines lists of query completions by demoting previously observed query completions based on the time they have been displayed to the user, allowing more relevant suggestions to be prominently displayed as the user continues typing, thereby improving the accuracy of query completions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If query completions are provided based on initial typing without dynamic refinement, then users receive a broad set of completion options, but the relevance and suitability of these completions deteriorates as users continue typing without feedback
Solution Approach 1:
The query completion system dynamically adjusts the list of suggested completions as users continue typing, rather than providing a static list. The system monitors user interactions (scrolling, hovering, selection) and updates the completion suggestions in real-time, making the system adaptive and responsive to user behavior patterns throughout the query formulation process
Solution Approach 2:
The system incorporates implicit feedback mechanisms by tracking user interactions with the completion list (such as scrolling behavior, hover duration, and selection actions). This feedback is used to refine and re-rank completion suggestions, allowing the system to learn from user preferences and improve the relevance of future completions without requiring explicit user input
2Ease of operation
If the same query completions are repeatedly displayed to users, then the system maintains consistency in suggestions, but user engagement and discovery of relevant alternatives deteriorates
Solution Approach 1:
The system changes the parameters of completion suggestions based on user interaction history and current query context. By adjusting ranking weights, filtering criteria, and suggestion diversity parameters dynamically, the system presents varied yet relevant completions that adapt to user behavior patterns while maintaining operational ease
3Measurement precision
If query completion lists are not dynamically refined, then the system maintains simple processing logic, but the accuracy and usefulness of completions deteriorates during extended query entry
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing user interaction data (scrolling patterns, hover behaviors, selection frequencies) as users interact with completion lists. This pre-collected data is then used to efficiently refine and re-rank completions in subsequent interactions, achieving high accuracy without requiring complex real-time analysis during each query entry session
Data Source
AI summary
Methods, systems, and apparatus are described that include receiving a sequence of one or more characters representing a first partial query entered into a search field on a computing device. A first list of query completions for the first partial query is provided for display on the computing device. One or more additional characters are then entered into the search field, where the sequence of characters and the additional characters cumulatively represent a second partial query. A second list of query completions are then obtained for the second partial query, and one or more query completions appearing in the first and the second lists are identified. Demotion scores for the identified query completions are then calculated which are used to demote the identified query completions to a lesser position within the second list, thereby forming a refined second list for display on the computing device.