Picker Expertise Tag Selection for Order-Specific User Display
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online concierge systems struggle to programmatically select the most relevant picker expertise tags to present to users, given the large dataset of picker information and user responses, which affects user satisfaction and engagement.
Innovation Solution
An online concierge system uses a trained model to select picker expertise tags based on user and order characteristics, optimizing performance metrics such as engagement rate and user tips, and presents these tags in a carousel interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all picker expertise information is presented to users, then user confidence and information completeness improve, but screen space and interface complexity worsen
Solution Approach 1:
The patent extracts only the most relevant picker expertise tags from the complete set of picker credentials and presents them to users. A trained model identifies and extracts the subset of expertise tags that are most relevant to the specific order context, thereby presenting essential information without overwhelming the limited screen space.
Solution Approach 2:
The patent applies local quality by tailoring the presented expertise information to the specific context of each order. Different users receiving different orders see different relevant expertise tags based on the order characteristics (e.g., international items, produce, meat), ensuring that the information displayed is locally optimized for each user-order context rather than presenting uniform all expertise information.
2Measurement precision
If a trained model is used to select picker expertise tags, then relevance and user satisfaction improve, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-training a model on historical data containing picker expertise information and user responses. This training phase occurs beforehand, allowing the system to learn patterns and relationships in the data. During actual order fulfillment, the pre-trained model can quickly select relevant expertise tags without requiring complex real-time computations, thus reducing operational system complexity while maintaining high relevance accuracy.
3Productivity
If picker expertise tags are selectively presented, then user engagement and satisfaction improve, but information loss increases
Solution Approach 1:
The patent incorporates feedback mechanisms by utilizing historical user response data during the model training phase. The system learns from past user interactions with picker expertise information, analyzing which expertise tags most effectively improve user engagement and satisfaction. This feedback loop enables the model to continuously improve its selection accuracy, presenting the right expertise information that maximizes user engagement while minimizing unnecessary information loss.
Data Source
AI summary
An online concierge system selects picker expertise tags that showcase abilities or experiences of pickers that fulfill orders for the system. The online concierge system establishes a set of user-order cohorts based on characteristics of orders and users placing the orders. When an order is received, the online concierge system identifies a relevant user-order cohort and applies a trained model to predict, in the context of the user-order cohort, the performance of various candidate picker expertise tags applicable to the order. The trained model may be generated via a training and testing process in which different picker expertise tags are tested in the context of a user-order cohort, and performance metrics are observed to learn which picker expertise tags perform best in the context of a user-order cohort.