Online Concierge Category Taxonomy Recommendation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online concierge systems require users to navigate through extensive lists of items or provide multiple search queries to select specific items for orders, leading to increased time expenditure and reduced user interaction and order frequency due to limited item recommendations based on user preferences.
Innovation Solution
An online concierge system employs a taxonomy of items with varying levels of specificity, using a trained classification model to associate items with categories and leverage prior orders to recommend additional categories and items, simplifying the order creation process by displaying relevant items based on user selections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional online concierge systems display a wide range of items to users, then users have access to a large number of items, but limited information about various items is available for recommendation
Solution Approach 1:
The patent introduces category information as an intermediary between individual items and recommendation logic. Instead of relying solely on item-level cooccurrence data, the system uses category-level information to bridge gaps when item-specific data is insufficient, enabling recommendations even for items with limited historical order data
Solution Approach 2:
The patent adds a category dimension to the recommendation system. By organizing items into hierarchical categories and analyzing cooccurrence at the category level, the system creates an additional layer of information that complements item-level data, allowing recommendations to be generated from multiple dimensional perspectives
2Ease of operation
If conventional systems require users to navigate through long lists of items or provide multiple search queries, then specific items can be identified, but users spend a considerable amount of time selecting items
Solution Approach 1:
The system performs preliminary analysis of order data to identify frequently cooccurring item categories before users place orders. By pre-computing category cooccurrence statistics and maintaining a taxonomy structure, the system prepares recommendation data in advance, enabling rapid generation of personalized item suggestions without requiring users to navigate through extensive item lists
Solution Approach 2:
The recommendation system automatically generates personalized item suggestions based on users' own ordering patterns and category cooccurrence data. The system serves itself by leveraging aggregated order information to create tailored recommendations, reducing the need for users to manually search or specify multiple criteria
3Adaptability or versatility
If conventional systems provide item recommendations based on previous cooccurrences, then specific item recommendations can be made, but items relevant to user preferences may not be displayed
Solution Approach 1:
The patent introduces category-level analysis as an additional dimension to the recommendation system. By examining cooccurrence patterns at the category level rather than only at the item level, the system captures broader user preferences and contextual relationships, enabling more versatile recommendations that adapt to different user needs and ordering contexts
Solution Approach 2:
The category taxonomy structure serves multiple functions simultaneously: it organizes items for navigation, enables recommendation generation, and provides a framework for analyzing user preferences. This multi-functional approach allows the same categorical structure to support both broad category-based recommendations and specific item-level suggestions
Data Source
AI summary
An online concierge system maintains a taxonomy associating one or more specific items offered by a warehouse with a category. When the online concierge system receives a selection of an item from a user for inclusion in an order, the online concierge system determines a category including the selected item. From prior received orders, the online concierge system 102 identifies additional categories including one or more items included in various prior received orders. Based on cooccurrences of the category and the additional categories, the online concierge system generates scores for the additional categories. An additional category is selected based on the scores and specific items from the selected additional category are displayed via an interface for selection by the user.