Product Graph for Context-Aware Substitute Recommendations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of online retail systems face challenges in finding relevant product substitutes as existing systems rely on trial-and-error with search criteria, failing to consider contextual expectations beyond explicit keywords or images.
Innovation Solution
A method and system utilizing a machine learning module to construct a product graph, which determines substitutes by analyzing the context of user inquiries and projecting requested products onto this graph to find relevant alternatives based on attributes and relationships within the domain.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional keyword-based search is used, then the system is simple to operate, but the search accuracy and relevance of substitutes are poor
Solution Approach 1:
The patent introduces a product graph as an intermediary structure between user queries and product inventory. The graph contains entities (products, attributes, categories) and their relationships, allowing the system to traverse semantic connections rather than relying on simple keyword matching. This intermediary enables context-aware substitute recommendations by analyzing paths and relationships in the graph.
Solution Approach 2:
The patent replaces traditional mechanical keyword-matching search mechanisms with a machine learning-based graph traversal system. The ML module automatically constructs and queries the product graph, using learned relationships and contextual understanding to identify substitutes, rather than relying on manual keyword configuration and exact matching.
2Measurement precision
If contextual information is considered, then the relevance of substitutes improves, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by pre-construction of the product graph structure and pre-computation of entity relationships during off-peak times. The graph is built in advance with all entities, attributes, and relationships already established, so that during actual product search, the system only needs to traverse the pre-built graph rather than computing relationships in real-time, significantly reducing query processing time.
3Productivity
If manual trial-and-error search is required, then the system requires minimal processing power, but the user time and effort increase significantly
Solution Approach 1:
The patent implements self-service by enabling the system to automatically understand user intent and generate substitute recommendations without requiring manual refinement of search criteria. The machine learning module autonomously interprets the product graph to identify relevant substitutes based on the initial query, eliminating the need for users to iteratively adjust search parameters.
Data Source
AI summary
A system can include one or more processors and non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform a method for determining one or more substitutes for a product requested in an inquiry from a user. The method comprises one or more of the steps including receiving an inquiry from a user, which comprises a requested product; extracting one or more attributes of the requested product based on the inquiry; determining a context of the inquiry; and determining one or more substitutes for the requested product from an inventory database based on a distance between the requested product and each substitute of the one or more substitutes, as projected in a product graph relating to a domain of the requested product and the context of the inquiry. In many embodiments, the product graph relating to the domain is generated by a machine learning module configured to dynamically learn information about the domain from various sources, including one or more expert opinions, the inventory database, and/or public records; and dynamically build the product graph according to the information about the domain. An exemplary machine learning module can be configured to dynamically build the product graph by: creating one or more product categories in the domain; extracting one or more entity nodes from the one or more product categories based on the information about the domain; and build one or more connections, including an entity-entity connection between each pair of related entity nodes, a product-entity connection between each product node and each of one or more entity nodes associated with the each product node, and a product-product connection between each pair of related product nodes. Other embodiments are disclosed.