Graph Partitioning Using Unified Neighborhood Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for combining electronic database records of product attributes with point of sale information within the domain of graph theory often result in inefficiencies and duplication due to rare associations between different types of data, leading to suboptimal data associations.
Innovation Solution
The method involves generating a structure graph from transactions, retrieving attributes, creating a structure-attribute graph, and using a unified neighborhood matrix to partition the graph, thereby combining structural and attribute-based graph partitioning to establish new data associations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If items are clustered based on similarity or frequency using traditional dendrogram methods, then hierarchical organization is achieved, but duplication and inefficiencies occur when combining database records
Solution Approach 1:
The patent merges structural graph partitioning (based on transaction relationships) with attribute-based graph partitioning (based on item characteristics) into a unified graph partitioning approach. This integration allows simultaneous consideration of both purchase patterns and product attributes, eliminating the need for separate clustering operations that cause duplication and inefficiency when combining transaction records with attribute data.
2Adaptability or versatility
If separate clustering systems are used for transaction data and attribute data, then each data type can be organized independently, but new associations between different data types cannot be created
Solution Approach 1:
The patent creates a universal graph partitioning framework that handles both structural relationships (from transactions) and attribute characteristics (from item data) within a single system. The unified neighborhood matrix serves multiple functions by capturing both types of relationships, enabling the system to discover associations across different data types without requiring separate specialized clustering systems.
Data Source
AI summary
Systems and methods for performing structure and attribute based graph partitioning. An exemplary system can receive a plurality of transactions, with each transaction identifying a point of sale and an item exchanged by the transaction. The system can generate a structure graph of the plurality of transactions and receive a request to partition the structure graph into sub-categories. The system can then retrieve, from a database, attributes of each respective structure vertex in the structure vertices. Using those attributes and the structure graph, the system can generate a structure-attribute graph. The system can then generate a unified neighborhood matrix of the structure-attribute graph; and partition the structure graph based on the unified neighborhood matrix.