US-12619937-B1 - Apparatus and method for directed graph modification and simulation based on external data

Abstract

An apparatus and method for directed graph modification and simulation based on external data are disclosed. The apparatus includes at least a processor, and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive external data from one or more external data sources, retrieve a directed graph including at least two internal nodes and at least one internal directed edge, wherein each internal node represents internal data and the at least one internal directed edge represents a relationship between the at least two internal nodes, generate a data extrapolation of the external data as a function of at least a part of the at least two internal nodes and the at least one internal directed edge, and modify the directed graph as a function of the data extrapolation.

Inventors

• Geoff Woods
• Randall Joseph Ottinger

Assignees

• AI Leadership Labs, LLC

Dates

Publication Date

20260505

Application Date

20250821

Claims (20)

1. 1 . A method for directed graph modification and simulation based on external data, the method comprising: receiving, using at least a processor, external data from one or more external data sources, wherein the external data comprises structured and unstructured data; retrieving, using the at least a processor, a directed graph comprising at least two internal nodes and at least one internal directed edge, wherein each internal node represents internal data and the at least one internal directed edge represents a relationship between the at least two internal nodes; generating, using the at least a processor, a data extrapolation of the external data as a function of at least a part of the at least two internal nodes and the at least one internal directed edge; constructing a decision tree as a function of the internal data, external data and the data extrapolation; modifying, using the at least a processor, the directed graph as a function of the data extrapolation, wherein modifying the directed graph comprises: generating at least one external node representing the external data; and generating at least one external directed edge connecting the at least one external node and at least one of the at least two internal nodes as a function of the data extrapolation; simulating, using the at least a processor, a plurality of sequential actions as a function of the data extrapolation and the modified directed graph, wherein simulating the plurality of sequential actions comprises identifying an action among the plurality of sequential actions that has a highest function score on the directed graph based on a simulated outcome of the plurality of sequential actions, wherein each node associated with the decision tree is associated with at least one of: a decision point and a state transition and each edge associated with the decision tree is associated with an action of the plurality of sequential actions and wherein the processor is further configured, using a decision-tree-based simulation engine, to: evaluate a change in one or more graph-related metrics, in response to the evaluate a change, identify the action that has a highest function score in order to predict a downstream impact; and generating, using the at least a processor, a user interface comprising the action that has the highest function score.
2. 2 . The method of claim 1 , wherein modifying the directed graph comprises modifying the at least one internal directed edge between the at least two internal nodes as a function of the at least one external node and the at least one external directed edge.
3. 3 . The method of claim 1 , wherein modifying the directed graph comprises: determining a function datum of the data extrapolation as a function of a correspondence between the external data and at least a part of the at least two internal nodes and the at least one internal directed edge; and generating the at least one external directed edge as a function of the function datum.
4. 4 . The method of claim 3 , wherein determining the function datum comprises: converting the external data to an external embedding; converting the internal data to an internal embedding; and generating the function score of the function datum as a function of a similarity distance between the external embedding and the internal embedding.
5. 5 . The method of claim 4 , wherein modifying the directed graph comprises: determining a weighted value of the at least one external node as a function of the function score; and generating a visual representation of the at least one external node as a function of the weighted value.
6. 6 . The method of claim 3 , wherein generating the user interface comprises generating the user interface comprising a function heatmap comprising a visual representation of the modified directed graph, wherein each node of the modified directed graph is rendered with a visual intensity based on the function score.
7. 7 . The method of claim 1 , wherein modifying the directed graph comprises: generating an action datum of the data extrapolation as a function of the external data and the internal data of the at least two internal nodes; and generating the at least one external node as a function of the action datum.
8. 8 . The method of claim 7 , wherein determining the action datum comprises: generating an actionable prompt as a function of the action datum; and transmitting the actionable prompt to a downstream device.
9. 9 . The method of claim 1 , wherein generating the data extrapolation comprises generating the data extrapolation using an extrapolation machine-learning model that has been trained on one or more extrapolation training datasets comprising exemplary external data and exemplary internal data correlated to exemplary data extrapolations.
10. 10 . The method of claim 1 , wherein generating the data extrapolation comprises: identifying one or more external data features of the external data; identifying one or more internal data features of the internal data; and determining at least one internal node related to the external data as a function of the one or more external data features and the one or more internal data features.
11. 11 . An apparatus for directed graph modification and simulation based on external data, the apparatus comprising: at least a processor; and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to: receive external data from one or more external data sources, wherein the external data comprises structured and unstructured data; retrieve a directed graph comprising at least two internal nodes and at least one internal directed edge, wherein each internal node represents internal data and the at least one internal directed edge represents a relationship between the at least two internal nodes; generate a data extrapolation of the external data as a function of at least a part of the at least two internal nodes and the at least one internal directed edge; modify the directed graph as a function of the data extrapolation, wherein modifying the directed graph comprises: generating at least one external node representing the external data; and generating at least one external directed edge connecting the at least one external node and at least one of the at least two internal nodes as a function of the data extrapolation; and constructing a decision tree as a function of the internal data, external data and the data extrapolation; simulate a plurality of sequential actions as a function of the modified directed graph, wherein simulating the plurality of sequential actions comprises identifying an action among the plurality of sequential actions that has a highest function score within the modified directed graph based on a simulated outcome of the plurality of sequential actions, wherein each node associated with the decision tree is associated with at least one of: a decision point and a state transition and each edge associated with the decision tree is associated with an action of the plurality of sequential actions and wherein the processor is further configured, using a decision-tree-based simulation engine, to: evaluate a change in one or more graph-related metrics, in response to the evaluate a change, identify the action that has a highest function score in order to predict a downstream impact; and generate a user interface comprising the modified directed graph and the action that has the highest function score.
12. 12 . The apparatus of claim 11 , wherein modifying the directed graph comprises modifying the at least one internal directed edge between the at least two internal nodes as a function of the at least one external node and the at least one external directed edge.
13. 13 . The apparatus of claim 11 , wherein modifying the directed graph comprises: determining a function datum of the data extrapolation as a function of a correspondence between the external data and at least a part of the at least two internal nodes and the at least one internal directed edge; and generating the at least one external directed edge as a function of the function datum.
14. 14 . The apparatus of claim 13 , wherein determining the function datum comprises: converting the external data to an external embedding; converting the internal data to an internal embedding; and generating the function score of the function datum as a function of a similarity distance between the external embedding and the internal embedding.
15. 15 . The apparatus of claim 14 , wherein modifying the directed graph comprises: determining a weighted value of the at least one external node as a function of the function score; and generating a visual representation of the at least one external node as a function of the weighted value.
16. 16 . The apparatus of claim 13 , wherein generating the user interface comprises generating the user interface comprising a function heatmap comprising a visual representation of the modified directed graph, wherein each node of the modified directed graph is rendered with a visual intensity based on the function score.
17. 17 . The apparatus of claim 11 , wherein modifying the directed graph comprises: generating an action datum of the data extrapolation as a function of the external data and the internal data of the at least two internal nodes; and generating the at least one external node as a function of the action datum.
18. 18 . The apparatus of claim 17 , wherein determining the action datum comprises: generating an actionable prompt as a function of the action datum; and transmitting the actionable prompt to a downstream device.
19. 19 . The apparatus of claim 11 , wherein generating the data extrapolation comprises generating the data extrapolation using an extrapolation machine-learning model that has been trained on one or more extrapolation training datasets comprising exemplary external data and exemplary internal data correlated to exemplary data extrapolations.
20. 20 . The apparatus of claim 11 , wherein generating the data extrapolation comprises: identifying one or more external data features of the external data; identifying one or more internal data features of the internal data; and determining at least one internal node related to the external data as a function of the one or more external data features and the one or more internal data features.

Description

FIELD OF THE INVENTION The present invention generally relates to the field of directed graph modification and simulation. In particular, the present invention is directed to an apparatus and method for directed graph modification and simulation based on external data. BACKGROUND Directed graphs are widely used data structures for modeling relationships between entities. While traditional directed graphs are typically constructed from static datasets, there is growing interest in systems that allow directed graphs to evolve over time in response to new data. However, existing systems lack technical mechanisms for dynamically modifying directed graphs in real time based on continuous inflows of heterogeneous data, specifically structured and unstructured data originating from multiple, independent, and often inconsistent external sources. Traditional directed graph systems are typically constructed from curated, static datasets in which the relationships between nodes are predefined and infrequently altered. These systems do not support automatic incorporation of external signals in a way that maintains coherence, semantic relevance, and data fidelity across a continuously evolving graph structure. Furthermore, existing systems do not provide mechanisms for deduplication that operate in real time while preserving relevance. Deduplication of streaming data is technically complex due to the need for fast, in-memory comparison of incoming data elements against a dynamically updating repository of recently processed signals. This must be done without introducing latency or false positives that could lead to incorrect suppression of valid but novel data. Accordingly, there exists a need for a technical improvement of dynamically modifying directed graph data structures in real time using heterogeneous external data sources. SUMMARY OF THE DISCLOSURE In some aspects, the techniques described herein relate to an apparatus for directed graph modification and simulation based on external data, the apparatus including at least a processor, and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to receive external data from one or more external data sources, wherein the external data includes structured and unstructured data, retrieve a directed graph including at least two internal nodes and at least one internal directed edge, wherein each internal node represents internal data and the at least one internal directed edge represents a relationship between the at least two internal nodes, generate a data extrapolation of the external data as a function of at least a part of the at least two internal nodes and the at least one internal directed edge, modify the directed graph as a function of the data extrapolation, wherein modifying the directed graph includes generating at least one external node representing the external data, and generating at least one external directed edge connecting the at least one external node and at least one of the at least two internal nodes as a function of the data extrapolation, simulate a plurality of sequential actions as a function of the modified directed graph, wherein simulating the plurality of sequential actions includes identifying an action among the plurality of sequential actions that has a highest function score within the modified directed graph based on a simulated outcome of the plurality of sequential actions, and generate a user interface including the modified directed graph and the action that has the highest function score. In some aspects, the techniques described herein relate to a method for directed graph modification and simulation based on external data, the method including receiving, using at least a processor, external data from one or more external data sources, wherein the external data includes structured and unstructured data, retrieving, using the at least a processor, a directed graph including at least two internal nodes and at least one internal directed edge, wherein each internal node represents internal data and the at least one internal directed edge represents a relationship between the at least two internal nodes, generating, using the at least a processor, a data extrapolation of the external data as a function of at least a part of the at least two internal nodes and the at least one internal directed edge, modifying, using the at least a processor, the directed graph as a function of the data extrapolation, wherein modifying the directed graph includes generating at least one external node representing the external data, and generating at least one external directed edge connecting the at least one external node and at least one of the at least two internal nodes as a function of the data extrapolation, simulating, using the at least a processor, a plurality of sequential actions as a function of the data extrapolation and the modified directed graph,