Gesture-Based Query Abandonment Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring query abandonment in search engines do not effectively differentiate between 'good' and 'bad' abandonment, relying on implicit feedback signals like clicks and dwell time, which are inadequate for assessing user satisfaction.
Innovation Solution
A system and method that utilize gesture movement data from user interactions to identify 'good' and 'bad' abandonment by extracting feature data from user feedback signals, which are then used to train predictive models to determine user satisfaction and improve search result relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If implicit feedback signals (clicks, dwell time) are used to measure query abandonment, then the measurement approach is simple, but the ability to differentiate between good and bad abandonment is insufficient
Solution Approach 1:
The patent introduces gesture data as an intermediary signal between user interaction and abandonment classification. Gesture data captures fine-grained movement patterns that serve as a mediator to differentiate between good and bad abandonment, bridging the gap between simple click data and complex satisfaction assessment
Solution Approach 2:
The patent changes the measurement parameters from coarse-grained implicit feedback (clicks, dwell time) to fine-grained gesture parameters (movement patterns, swipe directions, touch duration). This parameter transformation enables more precise differentiation of abandonment types while maintaining computational feasibility
2Measurement precision
If gesture data is collected and processed to identify good and bad abandonment, then the user satisfaction assessment improves, but the data processing complexity increases
Solution Approach 1:
The patent extracts specific feature data from gesture data that is most relevant to abandonment classification. By selecting only the most discriminative gesture features rather than processing all raw gesture data, the system achieves high measurement precision while controlling processing complexity
Solution Approach 2:
The patent segments the gesture data processing into distinct stages: data collection, feature extraction, and model application. This segmentation allows each stage to be optimized independently, improving satisfaction measurement while managing overall system complexity
3Adaptability or versatility
If traditional implicit feedback signals are used, then the system complexity is low, but the query abandonment classification capability is insufficient
Solution Approach 1:
The patent makes the feedback signal processing system multi-functional by using gesture data to serve multiple purposes: detecting user intent, classifying abandonment types, and assessing satisfaction. This universal approach enhances classification capability while consolidating processing functions rather than adding separate systems
Data Source
AI summary
Examples of the present disclosure describe systems and methods of identifying good and bad abandonment using gesture movement. In aspects, user feedback signals may be received by a client device in response to the viewing and/or navigation of query results. The feedback signals may be provided to a framework for determining and/or analyzing query abandonment. The framework may identify gesture data in the feedback signals and extract feature data from the gesture data. The feature data may be provided to a metrics component to determine metrics and/or satisfaction values for the feature data. The metrics and/or feature data may be used to train a predictive model to identify good abandonment in query results.