Graph database implemented knowledge mesh
A method for creating a graph database implemented knowledge mesh is disclosed. The method includes receiving, by a computer system, data from a plurality of different streams and identifying, by the computer system executing at least one machine learning model, a plurality of triples included in the data. The method also comprises filtering, by the computer system, the plurality of triples to identify a relevant subset of triples by applying an ontological filter, applying, by the computer system, a disambiguation routine to the relevant subset of triples to correlate entities included in the relevant subset to other entities in a graph database and determine a degree of confidence with each correlation, and creating, by the computer system, a knowledge mesh using the graph database. The knowledge mesh comprises the relevant subset of triples, each correlation identified by application of the disambiguation routine, and the degree of confidence with each correlation.
1. A method for updating a graph database implemented knowledge mesh, comprising:
receiving, by a computer system, data of a plurality of different content types from a plurality of different sources, wherein the data is enhanced by the inclusion of added attributes;
after the data is enhanced by the inclusion of added attributes, identifying, by the computer system executing at least one machine learning (ML) model, a plurality of triples included in the data;
filtering, by the computer system, the plurality of triples to identify a relevant subset of triples by applying an ontological filter to the plurality of triples;
determining, by the computer system, where to place each triple of the relevant subset of triples in a knowledge mesh data stored in a graph database, wherein the knowledge mesh data comprises a plurality of dimensions, and wherein the determining where to place each triple in the knowledge mesh data comprises reducing the dimensionality of the knowledge mesh data by:
selecting a single dimension of the plurality of dimensions corresponding to the triple and disregarding remaining non-selected dimensions of the plurality of dimensions; and
identifying potential candidates related to each entity in the triple by parsing the selected single dimension without considering the remaining non-selected dimensions;
applying, by the computer system, a disambiguation routine, that evaluates the plurality of dimensions, to the relevant subset of triples to correlate entities included in the relevant subset of triples to other entities in the graph database including the potential candidates and determine a degree of confidence with each correlated entry; and
updating, by the computer system, the knowledge mesh data in the graph database to include the relevant subset of triples, each correlation identified by application of the disambiguation routine, and the degree of confidence with each correlated entity.
2. The method of claim 1 , wherein the ML model comprises at least one of a noun-verb-noun extraction or a subject-predicate-object ML extraction.
3. The method of claim 2 , wherein the data is received from a plurality of different streams, and wherein the plurality of different streams comprises one or more of a text stream, an audio stream, a video stream, an image stream, or a custom stream.
4. The method of claim 3 , wherein at least some of the plurality of dimensions correspond to a different stream of data, a different data source, or a different type of source material.
5. The method of claim 4 , wherein the knowledge mesh data comprises intra-dimensional relationships between entities represented in a single dimension and interdimensional relationships between entities represented in different dimensions.