Personalized Digital Carousel Ranking via User-Item Affinity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing digital recommendation systems on websites, such as e-commerce platforms, often present item carousels in a static and non-personalized manner, leading to irrelevant recommendations for users, which can result in lost sales as customers may find recommendations uninteresting or embarrassing, causing them to abandon purchases and switch to other websites.

Innovation Solution

Implement a system that determines personalized digital carousels by calculating user-carousel scores based on historical user transactions and engagement data, using metrics like user-item affinity scores, user-carousel prior scores, and user-category discovery scores to rank and order carousels dynamically for each user, ensuring more relevant item recommendations are displayed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If carousels are displayed in a static and global manner for all users, then the system complexity is reduced and ease of operation is improved, but the relevance of recommendations to individual user preferences deteriorates

Engineering Contradiction:

Improveease of displaying carouselsVSAvoidadaptability to user preferences

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic carousel ordering by calculating user-carousel scores that adapt to individual user preferences and behaviors. The system dynamically determines the order of carousels based on real-time user data including transaction history, engagement metrics, and category discovery scores, transforming the static display into an adaptive, user-specific experience.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary calculations of user-carousel scores by analyzing historical transaction data, engagement data, and category discovery patterns before presenting carousels to users. This advance preparation allows the system to pre-compute relevance scores and optimize carousel ordering based on predicted user preferences, reducing real-time computational burden while maintaining personalization.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If personalized carousels are implemented based on user preferences and behaviors, then the relevance of recommendations is improved, but the computational complexity and processing requirements increase

Engineering Contradiction:

Improverelevance of recommendationsVSAvoidcomplexity of recommendation system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent segments the recommendation system into distinct computational components: user-item affinity scoring, carousel prior scoring, category discovery scoring, and final user-carousel score aggregation. Each component processes specific aspects of user behavior independently, allowing for modular implementation and optimization while maintaining overall system reliability and recommendation relevance.

Inventive Principle:

Principle #1Segmentation

3Measurement precision

If the system analyzes extensive user data including transactions and engagement metrics, then the accuracy of personalized recommendations is improved, but the time required to process and present carousels increases

Engineering Contradiction:

Improveprecision of user preference measurementVSAvoidtime to process user data

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary analysis of user transaction data and engagement metrics to pre-compute user-item affinity scores, carousel prior scores, and category discovery scores. By conducting these computations in advance before users interact with the system, the patent reduces real-time processing requirements while maintaining high measurement precision for personalized recommendations.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11308543B1Methods and apparatus for automatically providing personalized carousels

Publication Date: 2022.04.19 WALMART APOLLO LLC

AI summary

This application relates to apparatus and methods for automatically determining and providing carousels specifically curated for a user. In some examples, a computing device obtains user transaction data identifying in-store and/or online transactions, and user engagement data identifying user interactions with items and carousels from user's prior sessions. The computing device determines a sequential order for presentation of carousels with a set of item recommendations. For example, the computing device scores each potential carousel based on prior user interactions and transactions with items and carousels. The carousels are then ranked and subsequently presented to the user based on their corresponding scores.