Relationship Graph Network Pruning via Core Degree Mining

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing scale of user data on network platforms leads to high computing resource consumption and low data processing efficiency in big data analysis, necessitating a solution to improve analysis efficiency and reduce costs.

Innovation Solution

A data processing method that involves acquiring a relationship graph network, performing core degree mining through a device cluster, iteratively updating node core degrees, pruning the network based on core degrees, and compressing the device cluster when the network scale meets a compression condition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If distributed computing is performed through a device cluster for core degree mining, then data processing capability is improved, but computing resource consumption increases and time cost for data distribution increases

Engineering Contradiction:

Improvedata processing efficiencyVSAvoidcomputing resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent segments the device cluster into multiple computing devices that process different partitions of the relationship graph network independently. Each computing device processes a specific partition in parallel, improving overall data processing efficiency while reducing the computational burden on individual devices and the total resource consumption through distributed processing.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by first determining node core degrees and identifying pruning targets before actually pruning the relationship graph network. This preliminary analysis allows the system to plan the pruning operations in advance, reducing the time cost during the actual pruning process and optimizing resource utilization by processing only the necessary portions of the data.

Inventive Principle:

Principle #10Preliminary action

2Speed

If the relationship graph network is pruned by removing nodes and edges, then data scale is reduced and processing speed is improved, but data completeness may be compromised

Engineering Contradiction:

Improveprocessing speedVSAvoiddata completeness

Core Design Contradiction:

SpeedVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism where node core degrees are iteratively updated after each pruning operation. The system continuously monitors the core degrees of remaining nodes and adjusts pruning decisions based on this feedback, ensuring that pruning operations maintain data completeness by preserving nodes that are still important to the overall network structure while removing only those that can be safely pruned.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary determination of node core degrees and identification of pruning targets before executing the actual pruning. This preliminary action allows the system to carefully select which nodes and edges to remove based on their importance metrics, ensuring that data completeness is maintained by preserving critical nodes while still achieving processing speed improvements through selective pruning.

Inventive Principle:

Principle #10Preliminary action

3Loss of energy

If computing resources are reduced by compressing the device cluster, then cost is reduced, but data processing capacity decreases

Engineering Contradiction:

Improvecomputing resource consumptionVSAvoiddata processing capacity

Core Design Contradiction:

Loss of energyVSProductivity

Solution Approach 1:

The patent segments the relationship graph network into multiple partitions that can be processed independently by different computing devices. This segmentation allows the system to use fewer computing resources by processing only relevant partitions in parallel, reducing computing resource consumption while maintaining data processing capacity through efficient distributed processing of the segmented data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by selectively processing only the partitions and nodes that are most relevant to the current analysis task. Rather than processing the entire relationship graph network with all computing resources, the system identifies and processes only the necessary portions, reducing computing resource consumption while maintaining adequate data processing capacity for the specific analytical goals.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12216629B2Data processing method and apparatus, computerreadable medium, and electronic device

Publication Date: 2025.02.04 TENCENT TECHNOLOGY (SHENZHEN) CO LTD

AI summary

This application belongs to the technical field of artificial intelligence, and particularly relates to a data processing method, a data processing apparatus, a computer-readable medium, and an electronic device. The method includes acquiring a relationship graph network, the relationship graph network comprising nodes for representing interacting objects and edges for representing interactive relationships between multiple interacting objects; performing core degree mining on the relationship graph network through a device cluster comprising multiple computing devices, and iteratively updating node core degrees of all or some of the nodes in the relationship graph network; pruning the relationship graph network according to the node core degrees to remove some of the nodes and edges in the relationship graph network; and compressing the device cluster to remove some of the computing devices in the device cluster, when a network scale of the relationship graph network satisfies a network compression condition.