Granted Patent B1

US 12,639,757 · App. 18/963,020 · Granted May 26, 2026

Inventors: Brandon Krull (Santa Ana, CA); Syed M. Amir Husain (Georgetown, TX); Steven Lau (Westport, CT); Thiam Hui Lee (New York, NY); Aldo Marini Macouzet (Alameda, CA)

Assignee: ALPHA DEAL LLC

G06Q40/04G06N20/00G06Q40/0421G06Q40/06G06Q40/042

Abstract

Claims (78)

1 . A computer-implemented method executed by a controller of an integrated entity analysis system for structuring, indexing, retrieving, and analyzing entity data for automating asset trading, the method comprising:

for each entity of a plurality of entities:

configuring, by the controller, an AI model to generate one or more natural-language search strings based on information missing from an entity profile of the entity;

retrieving, by the controller, unstructured external data related to the entity from unstructured data sources using the natural-language search strings;

applying, by the controller, the unstructured external data to the AI model to generate structured property/value pairs describing the characteristics of the entity;

classifying, by the controller using the AI model, the entity based on the structured property/value pairs;

selecting, by the controller, an entity-profile template associated with the classification;

mapping, by the controller, the structured property/value pairs into corresponding fields of the entity-profile template and, when a property lacks a corresponding field, dynamically generating a new field in the entity profile and populating the new field with the corresponding value;

storing, by the controller, the updated entity profile in an entity database; and

converting, by the controller, the updated entity profile into a vector embedding and maintaining the vector embedding in a structured retrieval dataset comprising vector embeddings for a plurality of entity profiles;

receiving from a user, by the controller, investor-specified risk parameters and portfolio criteria;

converting, by the controller, a scoring-attribute definition or a user query into a vector embedding, and performing a similarity search across the structured retrieval dataset to identify portions of the entity profiles relevant to the scoring-attribute definition or the user query;

supplying, by the controller, the identified relevant portions of the entity profiles as context to the AI model, and generating, by the AI model, scoring data for the plurality of entity profiles based on the context;

based on scoring preferences of the user, selecting, by the controller, from the plurality of entity profiles using the scoring data, a first set of entity profiles corresponding to candidate assets that satisfy the investor-specified risk parameters and the portfolio criteria;

constructing, by the controller, a portfolio based on the investor-specified risk parameters, including the risk tolerance of the investor, and the portfolio criteria, the constructing including, for each candidate asset associated with an entity profile in the first set of entity profiles, determining based on the investor-specified risk parameters and the portfolio criteria, by the controller, an asset allocation for the portfolio; and

transmitting, by the controller, instructions to one or more trade fulfillment backends to perform one or more trades in accordance with the determined asset allocations,

wherein generating the structured property/value pairs, dynamically generating fields in the entity profile, converting the entity profiles into vector embeddings, maintaining the structured retrieval dataset, and performing similarity searches to supply context to the AI model reduces processing complexity, reduces computational overhead, improves scalability, and reduces energy consumption in the integrated entity analysis system.

2 . The method of claim 1 , wherein the one or more trade fulfillment backends include one or more blockchain platforms or trading services.

3 . The method of claim 1 , wherein determining the asset allocation for the portfolio includes utilizing linear programming, genetic algorithms, or machine learning models, to analyze the entity profiles and construct an optimized portfolio.

4 . The method of claim 3 further comprising continuously updating, by the controller, financial parameters for each asset in the optimized portfolio based on real-time market data, including changes in asset prices, market volatility, and sector performance.

5 . The method of claim 4 further comprising dynamically adjusting, by the controller, the asset allocations in the optimized portfolio in response to updated financial parameters.

6 . The method of claim 1 further comprising periodically recalculating, by the controller, asset valuations, risk assessments, and projected returns within the portfolio by incorporating historical portfolio performance data.

7 . The method of claim 1 further comprising verifying, by the controller, selected assets and trading activities comply with relevant regulatory requirements before transmitting the instructions to the one or more trade fulfillment backends to perform the one or more trades in accordance with the determined asset allocations.

8 . The method of claim 1 further comprising cross-referencing, by the controller, each trade against a database of regulatory requirements specific to the geographical regions and sectors of the selected assets, ensuring that all trades meet jurisdiction-specific legal standards.

9 . The method of claim 1 further comprising generating, by the controller, audit trails and compliance reports for each executed trade including recording transaction details, compliance checks, and regulatory confirmations.

10 . The method of claim 1 , further comprising providing, by the controller, simulation capabilities within a user interface including allowing users to model different portfolio scenarios based on hypothetical changes to risk parameters, asset allocations, or market conditions.

11 . The method of claim 1 further comprising:

for each asset associated with an entity profile in the first set of entity profiles, computing, by the controller, financial parameters for the asset including at least one of Sharpe ratio metrics, alpha metrics, and beta metrics; and

using, by the controller, the financial parameters to evaluate asset suitability and optimize the risk-adjusted return of the portfolio.

12 . An apparatus for automating asset trading based on integrated entity analysis, the apparatus comprising:

a processor;

one or more computer-readable storage media coupled to the processor; and

program instructions stored on the one or more storage media to cause the processor to perform operations comprising:

for each entity of a plurality of entities:

configuring, by the controller, an AI model to generate one or more natural-language search strings based on information missing from an entity profile of the entity;

retrieving, by the controller, unstructured external data related to the entity from unstructured data sources using the natural-language search strings;

applying, by the controller, the unstructured external data to the AI model to generate structured property/value pairs describing the characteristics of the entity;

classifying, by the controller using the AI model, the entity based on the structured property/value pairs:

selecting, by the controller, an entity-profile template associated with the classification:

mapping, by the controller, the structured property/value pairs into corresponding fields of the entity-profile template and, when a property lacks a corresponding field, dynamically generating a new field in the entity profile and populating the new field with the corresponding value:

storing, by the controller, the updated entity profile in an entity database; and

converting, by the controller, the updated entity profile into a vector embedding and maintaining the vector embedding in a structured retrieval dataset comprising vector embeddings for a plurality of entity profiles:

receiving from a user, by the controller, investor-specified risk parameters and portfolio criteria;

converting, by the controller, a scoring-attribute definition or a user query into a vector embedding, and performing a similarity search across the structured retrieval dataset to identify portions of the entity profiles relevant to the scoring-attribute definition or the user query:

supplying, by the controller, the identified relevant portions of the entity profiles as context to the AI model, and generating, by the AI model, scoring data for the plurality of entity profiles based on the context: based on scoring preferences of the user, selecting, by the controller, from the plurality of entity profiles using the scoring data, a first set of entity profiles corresponding to candidate assets that satisfy the investor-specified risk parameters and the portfolio criteria;

constructing, by the controller, a portfolio based on the investor-specified risk parameters, including the risk tolerance of the investor, and the portfolio criteria, the constructing including, for each candidate asset associated with an entity profile in the first set of entity profiles, determining based on the investor-specified risk parameters and the portfolio criteria, by the controller, an asset allocation for the portfolio; and

transmitting, by the controller, instructions to one or more trade fulfillment backends to perform one or more trades in accordance with the determined asset allocations;

wherein generating the structured property/value pairs, dynamically generating fields in the entity profile, converting the entity profiles into vector embeddings, maintaining the structured retrieval dataset, and performing similarity searches to supply context to the AI model reduces processing complexity, reduces computational overhead, improves scalability, and reduces energy consumption in the integrated entity analysis system.

13 . The apparatus of claim 12 , wherein the one or more trade fulfillment backends include one or more blockchain platforms or trading services.

14 . The apparatus of claim 12 , wherein determining the asset allocation for the portfolio includes utilizing linear programming, genetic algorithms, or machine learning models, to analyze the entity profiles and construct an optimized portfolio.

15 . The apparatus of claim 14 further comprising program instructions stored on the one or more storage media to cause the processor to perform operations comprising:

continuously updating, by the controller, financial parameters for each asset in the optimized portfolio based on real-time market data, including changes in asset prices, market volatility, and sector performance.

16 . The apparatus of claim 15 further comprising program instructions stored on the one or more storage media to cause the processor to perform operations comprising:

dynamically adjusting, by the controller, the asset allocations in the optimized portfolio in response to updated financial parameters.

17 . The apparatus of claim 12 further comprising program instructions stored on the one or more storage media to cause the processor to perform operations comprising: periodically recalculating, by the controller, asset valuations, risk assessments, and projected returns within the portfolio by incorporating historical portfolio performance data.

18 . The apparatus of claim 12 further comprising program instructions stored on the one or more storage media to cause the processor to perform operations comprising:

verifying, by the controller, selected assets and trading activities comply with relevant regulatory requirements before transmitting the instructions to the one or more trade fulfillment backends to perform the one or more trades in accordance with the determined asset allocations.

19 . The apparatus of claim 12 further comprising program instructions stored on the one or more storage media to cause the processor to perform operations comprising:

cross-referencing, by the controller, each trade against a database of regulatory requirements specific to the geographical regions and sectors of the selected assets, ensuring that all trades meet jurisdiction-specific legal standards.

20 . A computer program product for automating asset trading based on integrated entity analysis, the computer program product comprising:

a set of one or more computer readable storage media; and

computer program instructions, collectively stored in the set of one or more storage media, that when executed, cause a processor to perform computer operations comprising:

for each entity of a plurality of entities:

configuring, by the controller, an AI model to generate one or more natural-language search strings based on information missing from an entity profile of the entity;

retrieving, by the controller, unstructured external data related to the entity from unstructured data sources using the natural-language search strings;

applying, by the controller, the unstructured external data to the AI model to generate structured property/value pairs describing the characteristics of the entity;

classifying, by the controller using the AI model, the entity based on the structured property/value pairs;

selecting, by the controller, an entity-profile template associated with the classification;

mapping, by the controller, the structured property/value pairs into corresponding fields of the entity-profile template and, when a property lacks a corresponding field, dynamically generating a new field in the entity profile and populating the new field with the corresponding value;

storing, by the controller, the updated entity profile in an entity database; and

converting, by the controller, the updated entity profile into a vector embedding and maintaining the vector embedding in a structured retrieval dataset comprising vector embeddings for a plurality of entity profiles;

receiving from a user, by the controller, investor-specified risk parameters and portfolio criteria;

converting, by the controller, a scoring-attribute definition or a user query into a vector embedding, and performing a similarity search across the structured retrieval dataset to identify portions of the entity profiles relevant to the scoring-attribute definition or the user query;

supplying, by the controller, the identified relevant portions of the entity profiles as context to the AI model, and generating, by the AI model, scoring data for the plurality of entity profiles based on the context;

based on scoring preferences of the user, selecting, by the controller, from the plurality of entity profiles using the scoring data, a first set of entity profiles corresponding to candidate assets that satisfy the investor-specified risk parameters and the portfolio criteria;

constructing, by the controller, a portfolio based on the investor-specified risk parameters, including the risk tolerance of the investor, and the portfolio criteria, the constructing including, for each candidate asset associated with an entity profile in the first set of entity profiles, determining based on the investor-specified risk parameters and the portfolio criteria, by the controller, an asset allocation for the portfolio; and

transmitting, by the controller, instructions to one or more trade fulfillment backends to perform one or more trades in accordance with the determined asset allocations;

wherein generating the structured property/value pairs, dynamically generating fields in the entity profile, converting the entity profiles into vector embeddings, maintaining the structured retrieval dataset, and performing similarity searches to supply context to the AI model reduces processing complexity, reduces computational overhead, improves scalability, and reduces energy consumption in the integrated entity analysis system.

Assignments (1)

References Cited (170)