Network Graph
Network Graph is a Cortex-side job that constructs and analyzes network structures by generating scoped subgraphs and computing quantitative scores for nodes and edges. The job processes input parameters that define analytical boundaries and scope constraints, producing isolated subgraph representations focused on specific regions of interest within a larger network. This scoped approach enables targeted analysis without requiring full-graph processing, thereby reducing computational overhead.
Subgraph Generation and Scoping
The scoped subgraph generation component isolates portions of a larger network based on specified parameters. Rather than processing an entire network structure, the job generates subgraphs that represent defined boundaries or regions of analytical interest. This segmentation allows for more efficient computation and enables analysis to focus on relevant network sections without processing irrelevant data.
Node and Edge Scoring
Once subgraphs are generated, the job computes quantitative scores for both nodes and edges within those structures. These scores provide numerical representations of node and edge properties or relationships, supporting downstream analysis and decision-making. The scoring mechanism operates within the boundaries established by the scoped subgraph, ensuring that calculations remain focused on the relevant network portion.