Systems and methods for generating and implementing knowledge graphs for knowledge representation and analysis

A computer-implemented method comprising:

• building a knowledge graph having a plurality of subject nodes corresponding to the plurality of subjects and a plurality of loan nodes corresponding to attribute information;
• generating a hierarchical cluster tree structure by:
• identifying a first cluster of subject nodes associated with a first level of the hierarchical cluster tree structure;
• determining, for each subject node in the first cluster, at least one vector proportionate to a first loan edge weight associated with a respective property;
• identifying a first subcluster of subject nodes associated with the first cluster of the hierarchical cluster tree structure;
• determining, for each loan node in the first subcluster, at least one vector proportionate to a first neighbor edge weight associated with the respective subject node;
• and
• generating the hierarchical cluster tree structure based on the identified first cluster of subject nodes, the identified first subcluster of subject nodes, the determined at least one vector proportionate to the first loan edge weight, and the determined at least one vector proportionate to the first neighbor edge weight;
• determining, using a machine learning analysis, a geo-spatial neighborhood delineation based on the generated hierarchical cluster tree structure, wherein the machine learning analysis includes classification training through pruning noisy clusters through generalization;
• and
• updating, using the machine learning analysis, the knowledge graph by changing an identified neighborhood for one or more homes represented as nodes on the knowledge graph based on the determined geo-spatial neighborhood delineation.

A non-transitory computer readable medium storing instructions which, when executed, cause at least one processor to perform operations comprising:

• at least one processor;
• building a knowledge graph having a plurality of subject nodes corresponding to the plurality of subjects and a plurality of loan nodes corresponding to attribute information;
• generating a hierarchical cluster tree structure by:
• identifying a first cluster of subject nodes associated with a first level of the hierarchical cluster tree structure;
• determining, for each subject node in the first cluster, at least one vector proportionate to a first loan edge weight associated with a respective property;
• identifying a first subcluster of subject nodes associated with the first cluster of the hierarchical cluster tree structure;
• determining, for each loan node in the first subcluster, at least one vector proportionate to a first neighbor edge weight associated with the respective subject node;
• and
• generating the hierarchical cluster tree structure based on the identified first cluster of subject nodes, the identified first subcluster of subject nodes, the determined at least one vector proportionate to the first loan edge weight, and the determined at least one vector proportionate to the first neighbor edge weight;
• determining, using a machine learning analysis, a geo-spatial neighborhood delineation based on the generated hierarchical cluster tree structure, wherein the machine learning analysis includes classification training through pruning noisy clusters through generalization;
• and
• updating, using the machine learning analysis, the knowledge graph by changing an identified neighborhood for one or more homes represented as nodes on the knowledge graph based on the determined geo-spatial neighborhood delineation.