Explicit Associative Relationship Recommendation System

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing online shopping platforms struggle to accurately recommend items to consumers based on transactional and behavioral data, often failing to reflect the consumer's true interests due to the lack of explicit associative relationships between items.

Innovation Solution

Implementing a system that establishes and leverages explicit associative relationships between items and categories, using a controlled vocabulary to define relationships such as essential, accessory, or suggested items, and utilizing transactional data, consumer input, and merchant insights to recommend relevant products.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If recommendations are based on transactional data (items frequently purchased together), then the system can generate recommendations automatically, but the recommendations may not reflect true item relationships or consumer interests

Engineering Contradiction:

Improverecommendation generation efficiencyVSAvoiditem relationship accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary controlled vocabulary that mediates between raw transactional data and recommendation generation. This vocabulary serves as a bridge that transforms implicit purchase patterns into explicit, semantically meaningful item relationships, thereby improving accuracy without sacrificing automation efficiency

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary action by pre-defining a controlled vocabulary of item relationships before generating recommendations. This advance preparation creates a structured framework that guides subsequent recommendation generation, ensuring accuracy is built into the system architecture rather than added as a post-processing step

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If recommendations are based on behavioral data (viewing, search, navigation history), then the system can personalize recommendations, but the behavioral data may not accurately reflect consumer interests or preferences

Engineering Contradiction:

Improverecommendation personalizationVSAvoidconsumer interest accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements feedback mechanisms where consumer interactions with recommendations (clicks, purchases, ignores) are used to refine and update the controlled vocabulary of item relationships. This continuous feedback loop allows the system to adapt to changing consumer preferences while maintaining accuracy through structured relationship definitions

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system applies dynamics by making the controlled vocabulary adaptive and evolving based on observed consumer behavior patterns. Rather than static relationships, the item relationships are dynamically adjusted to reflect actual consumer interests, combining personalization with precision

Inventive Principle:

Principle #15Dynamics

3Measurement precision

If the system establishes explicit associative relationships between items using controlled vocabulary, then recommendation accuracy improves, but system complexity increases

Engineering Contradiction:

Improverecommendation accuracyVSAvoidsystem structure complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the complex task of recommendation generation into distinct components: a controlled vocabulary layer for defining item relationships, a data processing layer for analyzing transactional and behavioral data, and a recommendation generation layer for producing personalized suggestions. This segmentation reduces overall system complexity by breaking down the monolithic problem into manageable, independent modules

Inventive Principle:

Principle #1Segmentation

4Reliability

If the system provides detailed item relationships and recommendations, then consumers become more informed and purchase likelihood increases, but processing and storage requirements increase

Engineering Contradiction:

Improvepurchase decision confidenceVSAvoiddata processing volume

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent extracts only the essential and most relevant item relationships into the controlled vocabulary, rather than processing and storing all possible item associations. This selective extraction reduces data volume while maintaining the quality and reliability of recommendations by focusing on the most significant relationships

Inventive Principle:

Principle #2Taking out (Extraction)

Data Source

PatentUS9569419B1Associative relationship based recommendations

Publication Date: 2017.02.14 AMAZON TECH INC

AI summary

The systems and/or processes described herein may establish a controlled and limited vocabulary that may serve as explicit associative relationships. The explicit associative relationships may define the nature of relationships between items and/or categories of items. In response to determining that a user has interacted with or selected an item via a website, an application, etc., associated with a service provider, explicit associative relationships associated with the selected item may be parsed in order to identify additional items related to the selected item. The additional related items may then be dynamically recommended to the user via the website, the application, etc., associated with the service provider.