IP Library Granted Patent US 12,288,261
Granted Patent B2
US 12,288,261 · App. 18/963,512 · Granted Apr 29, 2025

Data retrieval and validation for asset onboarding

Inventors: Damien Patton (Plano, TX); Christian Gratton (Eaton Rapids, MI)
Assignee: TRETE Inc.
G06Q40/04G06F16/215G06F16/2358G06F16/2379G06Q20/363G06Q20/3674G06Q20/389G06Q20/401G06Q20/42G06Q30/0613G06Q40/06G06Q2220/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,288,261
App. No.
18/963,512
Filed
Nov 28, 2024
Granted
Apr 29, 2025
Kind
B2
Art Unit
2165
USPC
707/692
Abstract

In certain aspects of the disclosure, a computer-implemented method includes collecting, via an artificial intelligence module, a first set of data associated with an asset and creating identifiers associated with the first set of data. The method includes collecting a second set of data associated with the asset based on the identifiers and comparing the first set of data and the second set of data based on the identifiers. The method includes determining validation of the first set of data based on the comparison and generating a result of approval or rejection based on the comparison.

Claims (66)

1. A computer-implemented method for onboarding a real property asset into an Alternative Trading System (ATS), the method comprising:

electronically collecting a digital image, supplied by an owner of the real property asset and purported to be an image of the real property asset, into a database;

finding unique identifiers corresponding to each of one or more property features depicted within the digital image;

automatically and subsequent to accessing the unique identifiers, utilizing an artificial intelligence module validating relevancy of the depicted one or more depicted property features, including:

electronically collecting a second data set into the database using the unique identifiers, the second data set considered as corresponding to the real property asset based on the unique identifiers;

cross-referencing the digital image with the second data set; and

confirming the depicted one or more property features are features of the real property asset based on results of the cross-referencing;

responsive to the confirmation, listing a fractional interest in the real property asset for sale at the ATS; and

automatically, as part of a constant training cycle and concurrently with listing the fractional interest, training the artificial intelligence module using the confirmation, the depicted one or more property features, and the second data set as training data improving data validating performance of the artificial intelligence module.

2. The method of claim 1 , wherein electronically collecting a second data set into the database using the unique identifiers comprises collecting a second digital image known to be an image of the real estate asset from one of: the Internet, satellite imagery, or survey data;

wherein cross-referencing the digital image with the second data set comprises comparing the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein confirming the depicted one or more property features are features of the real property asset based on results of the cross-referencing comprises confirming the depicted one or more property features are also depicted in the second digital image.

3. The method of claim 1 , wherein finding unique identifiers corresponding to each of the depicted one or more property features within the digital image comprises creating an image fingerprint from the digital image;

wherein electronically collecting a second data set comprises collecting a second digital image using the image fingerprint;

wherein cross-referencing the digital image with the second data set comprises comparing the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein confirming the depicted one or more property features are features of the real property asset comprises confirming the digital image and the second digital image are both images of the real property asset based on the determined similarity.

4. The method of claim 1 , further comprising identifying an image descriptor within the image;

wherein finding unique identifiers corresponding to each of a plurality of relevant items comprises formulating a search tag from the image descriptor;

wherein electronically collecting a second data set comprises collecting a second digital image using the search tag;

wherein cross-referencing the digital image with the second data set comprises comparing the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein confirming the depicted one or more property features are features of the real property asset comprises confirming the digital image and the second digital image are both images of the real property asset based on the determined similarity.

5. The method of claim 1 , further comprising, prior to validating relevancy of the depicted one or more property features, an image classification algorithm auto-identifying a context within the digital image; and

wherein validating relevancy of the depicted one or more property features comprises validating relevancy of the depicted one or more property features to the real property asset based on the context.

6. The method of claim 1 , further comprising, prior to validating relevancy of the depicted one or more property features, an image quality algorithm auto-determining quality of the digital image and quality the second digital image are sufficient for cross-referencing.

7. The method of claim 1 , further comprising:

prior to collecting the digital image, electronically collecting an additional digital image, purported by the property owner to be an image of the real property asset, into the database; and

an image similarity algorithm auto-determining the digital image adds incremental value to the ATS by being sufficiently different from the additional digital image.

8. The method of claim 1 , wherein confirming the depicted one or more property features are features of the real property asset comprises a model:

scoring the cross-referencing by generating a confidence score representing a probability both the digital image and the second digital image are of the real property asset; and

determining the confidence score satisfies a control threshold.

9. The method of claim 1 , further comprising electronically collecting a third digital image from a different data source than the second digital image; and

wherein cross-referencing the digital image with the second digital image comprises cross-referencing the digital image with the second digital image and the third digital image.

10. A system comprising:

a processor;

system memory coupled to the processor and storing instructions configured to cause the processor to:

electronically collect a digital image, supplied by an owner of the real property asset and purported to be an image of the real property asset, into a database;

find unique identifiers corresponding to each of one or more property features depicted within the digital image;

automatically and subsequent to accessing the unique identifiers, utilize an artificial intelligence module validating relevancy of the depicted one or more property features, including:

electronically collect a second data set into the database using the unique identifiers, the second data set considered as corresponding to the real property asset based on the unique identifiers;

cross-reference the digital image with the second data set; and

confirm the depicted one or more property features are features of the real property asset based on results of the cross-referencing;

responsive to the confirmation, list a fractional interest in the real property asset for sale at the ATS; and

automatically, as part of a constant training cycle and concurrently with listing the fractional interest, train the artificial intelligence module using the confirmation, the depicted one or more property features, and the second data set as training data improving data validating performance of the artificial intelligence module.

11. The system of claim 10 , wherein instructions configured to cause the processor to electronically collect a second data set into the database using the unique identifiers comprise instructions configured to cause the processor to collect a second digital image known to be an image of the real estate asset from one of: the Internet, satellite imagery, or survey data;

wherein instructions configured to cause the processor to cross-reference the digital image with the second data set comprise instructions configured to cause the processor to compare the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein instructions configured to cause the processor to confirm the depicted one or more property features are features of the real property asset based on results of the cross-referencing comprise instructions configured to cause the processor to confirm the depicted one or more property features are also depicted in the second digital image.

12. The system of claim 10 , wherein instructions configured to cause the processor to find unique identifiers corresponding to each of the depicted one or more property features within the digital image comprise instructions configured to cause the processor to create an image fingerprint from the digital image;

wherein instructions configured to cause the processor to electronically collect a second data set comprise instructions configured to cause the processor to collect a second digital image using the image fingerprint;

wherein instructions configured to cause the processor to cross-reference the digital image with the second data set comprise instructions configured to cause the processor to compare the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein instructions configured to cause the processor to confirm the depicted one or more property features are features of the real property asset comprise instructions configured to cause the processor to confirm the digital image and the second digital image are both images of the real property asset based on the determined similarity.

13. The system of claim 10 , further comprising instructions configured to cause the processor to identify an image descriptor within the image;

wherein instructions configured to cause the processor to find unique identifiers corresponding to each of a plurality of relevant items comprise instructions configured to cause the processor to formulate a search tag from the image descriptor;

wherein instructions configured to cause the processor to electronically collect a second data set comprise instructions configured to cause the processor to collect a second digital image using the search tag;

wherein instructions configured to cause the processor to cross-reference the digital image with the second data set comprise instructions configured to cause the processor to compare the digital image to the second digital image determining a similarity between the digital image and the second digital image; and

wherein instructions configured to cause the processor to confirm the depicted one or more property features are features of the real property asset comprise instructions configured to cause the processor to confirm the digital image and the second digital image are both images of the real property asset based on the determined similarity.

14. The system of claim 10 , further comprising instructions configured to cause the processor to cause an image classification algorithm to, prior to validating relevancy of the depicted one or more property features, auto-identify a context within the digital image; and

wherein instructions configured to cause the processor to validate relevancy of the depicted one or more property features comprise instructions configured to cause the processor to validate relevancy of the depicted one or more property features to the real property asset based on the context.

15. The system of claim 10 , further comprising instructions configured to cause the processor to cause an image quality algorithm to, prior to validating relevancy of the depicted one or more property features, auto-determine quality of the digital image and quality the second digital image are sufficient for cross-referencing.

16. The system of claim 10 , further comprising instructions configured to cause the processor to:

prior to collecting the digital image, electronically collect an additional digital image, purported by the property owner to be an image of the real property asset, into the database; and

cause an image similarity algorithm to auto-determine the digital image adds incremental value to the ATS by being sufficiently different from the additional digital image.

17. The system of claim 10 , wherein instructions configured to cause the processor to confirm the depicted one or more property features are features of the real property asset comprise instructions configured to cause the processor to cause a model to:

score the cross-referencing by generating a confidence score representing a probability both the digital image and the second digital image are of the real property asset; and

determine the confidence score satisfies a control threshold.

18. The system of claim 10 , further comprising instructions configured to cause the processor to electronically collect a third digital image from a different data source than the second digital image; and

wherein instructions configured to cause the processor to cross-reference the digital image with the second digital image comprise instructions configured to cause the processor to cross-reference the digital image with the second digital image and the third digital image.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 19, 2025
From: PATTON, DAMIEN; GRATTON, CHRISTIAN
To: TRETE INC.
Reel/Frame 070258/0403 →
Continuity (14)
Continuation 18616149 · Mar 25, 2024
Provisional Application 63615128 · Dec 27, 2023
Provisional Application 63615108 · Dec 27, 2023
Provisional Application 63615145 · Dec 27, 2023
Provisional Application 63615136 · Dec 27, 2023
Provisional Application 63600381 · Nov 17, 2023
Provisional Application 63596471 · Nov 6, 2023
Provisional Application 63515337 · Jul 24, 2023
Provisional Application 63509257 · Jun 20, 2023
Provisional Application 63509266 · Jun 20, 2023
Provisional Application 63509264 · Jun 20, 2023
Provisional Application 63509261 · Jun 20, 2023
Provisional Application 63454622 · Mar 24, 2023
Related Publication 20250095071A1 · Mar 20, 2025
References Cited (46)
US 11989766B2 · Reynolds et al. · 2024 [cited by applicant]
US 11989776B1 · Hepp · 2024 [cited by examiner]
US 20160281607A1 · Asati · 2016 [cited by examiner]
US 20170161758A1 · Towriss · 2017 [cited by examiner]
US 20190171428A1 · Patton · 2019 [cited by applicant]
US 20190304104A1 · Amer · 2019 [cited by examiner]
US 20210034960A1 · Khapall · 2021 [cited by applicant]
US 20220067738A1 · Fang · 2022 [cited by applicant]
US 20220207349A1 · Fusco · 2022 [cited by applicant]
US 20220222657A1 · Nichani · 2022 [cited by applicant]
US 20220398337A1 · Franquin · 2022 [cited by examiner]
US 20230049167A1 · Wilson · 2023 [cited by applicant]
US 20230097897A1 · Campos · 2023 [cited by applicant]
US 20230148321A1 · Hall · 2023 [cited by applicant]
US 20230162088A1 · Woodward · 2023 [cited by applicant]
US 20230297831A1 · Saadi · 2023 [cited by applicant]
US 20240202339A1 · Fitzgerald · 2024 [cited by applicant]
US 20240220824A1 · Gilad · 2024 [cited by applicant]
US 20240235921A9 · Naga · 2024 [cited by examiner]
US 20240249290A1 · Mikhael · 2024 [cited by applicant]
US 20240257255A1 · Gupta · 2024 [cited by applicant]
US 20240340314A1 · Radon · 2024 [cited by applicant]
US 20240354436A1 · Mukherjee · 2024 [cited by applicant]
US 20240362716A1 · Woiwood · 2024 [cited by applicant]
CA 3062805A1 · 2020 [cited by applicant]
CA 3173084A1 · 2021 [cited by applicant]
WO 2019141984A1 · 2019 [cited by applicant]
WO 2023183494A1 · 2023 [cited by applicant]
Amoussou, Mandela, “What is a Cap Table, and Why are they Important?”, from Digital Securities, Published Jan. 10, 2023. [cited by applicant]
Politou, E. et al., “Blockchain Mutability: Challenges and Proposed Solutions”, from IEEE Transcations on Emerging Topics in Computing, Published Oct. 25, 2019. [cited by applicant]
Non-Final Office Action, U.S. Appl. No. 18/616,143, Mailing Date Aug. 8, 2024, 15 pages. [cited by applicant]
Notice of Allowance, U.S. Appl. No. 18/616,143, Maling Date Oct. 17, 2024, 8 pages. [cited by applicant]
Non-Final Office Action, U.S. Appl. No. 18/616,149, Mailing Date Jul. 17, 2024, 14 pages. [cited by applicant]
Final Office Action, U.S. Appl. No. 18/616,149, Mailing Date Oct. 15, 2024, 6 pages. [cited by applicant]
Notice of Allowance, U.S. Appl. No. 18/616,149, Mailing Date Nov. 12, 2024, 8 pages. [cited by applicant]
Non-Final Office Action, U.S. Appl. No. 18/620,299, Mailing Date Jul. 30, 2024, 13 pages. [cited by applicant]
Non-Final Office Action, U.S. Appl. No. 18/620,299, Mailing Date Nov. 27, 2024, 21 pages. [cited by applicant]
Li et al,. “TradingGPT: Multi-Agent System with Layered Memory and Distinct Characters for Enhanced Financial Trading Performance”, Sep. 7, 2023, arXiv:2309.03736v1, pp. 1-7. (Year: 2023). [cited by applicant]
Romanko et al., “ChatGPT-based Investment Portfolio Seleciton”, Aug. 11, 2023, arXiv:2308.06260, pp. 1-25. (Year: 2023). [cited by applicant]
Wu, Bingzhe “Is GPT4 a Good Trader?” Sep. 20, 2023, arXiv2309.10982v1, pp. 1-6. (Year: 2023). [cited by applicant]
Kang at al., “Deficiency of Large Language Models in Finance: An Emperical Examination of Hallucination”, Nov. 27, 2023, arXiv:2311.15548v1, pp. 1-15. (Year: 2023). [cited by applicant]
Crall, Jon, “The MCC Approaches the Geometric Mean of Precision and Recall as True Negatives Approach Infinity”, Jul. 9, 2023, arXiv: 2305.00594v2, pp. 1-4. (Year: 2023). [cited by applicant]
Kynkaaniemi et al., “Improved Precision and Recall For Assessing Generative Models”, Oct. 30, 2019, arXiv: 1904.06991v3, pp. 1-16. (Year:2019). [cited by applicant]
Sajjadi et al., “Assessing Generative Models via Precision and Recall”, Oct. 28, 2018, arXiv: 1806.00035v3. pp. 1-15. (Year: 2018). [cited by applicant]
Yang et al., “FinGPT: Open-Source Financial Large Language Models”, Jun. 9, 2023, arXiv: 2306.06031v1, pp. 1-7. (Year: 2023). [cited by applicant]
Zhang et al., “Enhancing Financial Sentiment Analysis via Retrieval Augmented Large Language Models”, Nov. 4, 2023, arXiv: 2310.04027v2, pp. 1-8. (Year: 2023). [cited by applicant]
Cited By (1)
US 12,597,002