Adaptive Query Suggestion Ranking With In-Session Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing query suggestion systems are static and do not adapt rankings based on user feedback during a search session, leading to ineffective query suggestion rankings.
Innovation Solution
A system that dynamically adjusts query suggestion rankings based on user interactions within a search session, using adaptive ranking scores calculated from user feedback such as dwell time, query selection, and typing patterns to promote or demote suggestions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If query suggestion rankings are pre-indexed and kept static throughout a search session, then the system maintains simple and fast query suggestion delivery, but the rankings fail to adapt to user feedback and user intent changes during the session
Solution Approach 1:
The patent implements dynamic query suggestion rankings that adapt during a search session based on user interactions. The system transitions from static pre-indexed rankings to dynamic rankings that are recalculated based on user feedback signals such as query modifications, selection behavior, and dwell time. This allows the query suggestion system to evolve and adapt to changing user intent throughout the session.
Solution Approach 2:
The patent incorporates user feedback mechanisms where user interactions with query suggestions (such as modifying the query, selecting a suggestion, or ignoring suggestions) are captured and used to adjust rankings. The system processes these feedback signals and uses them to promote or demote query suggestions in real-time, creating a closed-loop system that continuously improves based on user behavior.
2Measurement precision
If query suggestion rankings are dynamically adjusted based on user feedback during a search session, then the accuracy and relevance of query suggestions improve, but the computational complexity and processing time increase
Solution Approach 1:
The patent pre-calculates and stores baseline query suggestion rankings before the search session begins. These pre-indexed rankings serve as the initial foundation, allowing the system to quickly deliver suggestions without immediate computational overhead. During the session, only incremental adjustments are made based on user feedback, rather than complete recalculation, thus reducing processing time while maintaining accuracy.
Solution Approach 2:
The patent implements selective ranking adjustment where not all query suggestions are recalculated with full complexity. Instead, the system focuses computational resources on adjusting rankings for suggestions that receive user feedback, leaving unchanged the rankings of suggestions that are not interacted with. This partial action approach reduces overall processing time while still improving accuracy for relevant suggestions.
Data Source
AI summary
Methods, systems and programming for providing query suggestions based on user feedback. In one example, a prefix of a query is first received. An input including a prefix of a query is received from a user in a search session. A plurality of query suggestions are fetched based on the prefix of the query. Rankings of the plurality of query suggestions are determined based, at least in part, on the user's previous interactions in the search session with respect to at least one of the plurality of query suggestions. The at least one of the plurality of query suggestions has been previously provided to the user in the search session. The plurality of query suggestions are provided in the search session based on their rankings as a response to the input.