IP Library › Granted Patent US 12,547,398
Granted Patent B2
US 12,547,398 · App. 18/605,691 · Granted Feb 10, 2026

Systems and methods for managing a software repository

Inventors: Brian Burgess (Richmond, VA); Benjamin Cabell Glancy (Richmond, VA); Narender Pashikant (Ashburn, VA); Guganathan Sellamuthu (Glen Allen, VA)
Assignee: CAPITAL ONE SERVICES, LLC
G06F8/73G06F16/90332G06F16/90335G06N3/08
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 12,547,398
App. No.
18/605,691
Filed
Mar 14, 2024
Granted
Feb 10, 2026
Kind
B2
Examiner
VO, TED T
Art Unit
2191
USPC
717/120
Abstract

A system method of managing a software repository. The method including receiving a dataset comprising a set of computer instructions, retrieving feature-related data from the dataset, determining, by a machine learning model executing on the computing device, a function of the computer instructions based on the feature-related data, and generating a synopsis of the function of the computer instructions. The method can include transmitting the synopsis to a user interface and receiving feedback from a user indicative of whether the synopsis accurately describes the function of the computer instructions. If the feedback indicates that the synopsis accurately describes the function of the computer instructions, the method can include storing the synopsis in a memory of the computing device. If the feedback indicates that the synopsis inaccurately describes the function of the computer instructions, the method can include generating a revised synopsis of the function of the computer instructions.

Claims (61)

1 . A method of managing a software repository, the method comprising:

receiving a dataset comprising a set of computer instructions;

parsing, by a machine learning model, feature-related data from the dataset;

training the machine learning model using the feature-related data from the dataset;

determining, by the machine learning model using natural language processing, a function of the set of computer instructions based on text in comments included in the feature-related data;

generating, by the machine learning model, a synopsis comprising a description of the function of the set of computer instructions;

storing the synopsis in a memory of a device;

receiving feedback from a user indicative of whether the synopsis accurately describes the function of the set of computer instructions; and

in response to receiving the feedback, training the machine learning model with the feedback from the user.

2 . The method of claim 1 , further comprising:

in response to determining that the feedback indicates that the synopsis accurately describes the function of the set of computer instructions, storing the synopsis in the memory of the device; and

in response to determining the feedback indicates that the synopsis inaccurately describes the function of the set of computer instructions, generating, by the machine learning model, a revised synopsis of the function of the set of computer instructions based on the feedback from the user.

3 . The method of claim 2 , further comprising:

receiving a request to search the synopsis based on a search term; and

in response to determining that the set of computer instructions performs a function related to the search term, transmitting the synopsis to a user interface.

4 . The method of claim 2 , wherein generating the revised synopsis of the function of the set of computer instructions further comprises:

generating the revised synopsis based on the feedback from a user interface, the feedback comprising edits to the synopsis made by a user.

5 . The method of claim 1 , wherein generating the synopsis comprises natural language processing.

6 . The method of claim 1 , wherein in the machine learning model comprises a recurrent neural network.

7 . The method of claim 1 , wherein the feature-related data comprises programming language of the set of computer instructions.

8 . A method of managing a software repository, the method comprising:

parsing, by a machine learning model, feature-related data from a dataset;

training the machine learning model using the feature-related data from the dataset;

determining, by the machine learning model, a first function of a set of computer instructions based on the feature-related data;

generating, by the machine learning model, a synopsis comprising a description of the first function of the set of computer instructions;

storing the synopsis in a memory;

receiving a request to search the synopsis based on a search term, wherein the request comprises a second description of the first function for the set of computer instructions;

determining, based on the request, whether the set of computer instructions would perform a second function related to the search term;

transmitting the synopsis to a user interface;

receiving feedback from a user indicative of whether the synopsis accurately describes the first function of the set of the computer instructions; and

in response to receiving the feedback, training the machine learning model with the feedback from the user.

9 . The method of claim 8 , further comprising:

in response to determining that the feedback indicates that the synopsis accurately describes the first function of the set of computer instructions, storing the synopsis in the memory.

10 . The method of claim 9 , further comprising:

in response to determining the feedback indicates that the synopsis inaccurately describes the first function of the set of computer instructions, generating a revised synopsis of the first function of the set of computer instructions.

11 . The method of claim 10 , wherein generating a revised synopsis of the first function of the set of computer instructions further comprises:

generating the revised synopsis based on the feedback from a second user interface, the feedback comprising edits to the synopsis made by the user.

12 . The method of claim 8 , wherein generating the synopsis comprises natural language processing.

13 . The method of claim 8 , wherein the machine learning model comprises a recurrent neural network.

14 . The method of claim 8 , wherein the feature-related data comprises programming language of the set of computer instructions.

15 . A system for managing a software repository, the system comprising:

one or more processors; and

a memory in communication with the one or more processors and storing instruction that are configured to cause the system to:

parse, by a machine learning model executing on the one or more processors, feature-related data from a dataset, the feature-related data comprising data indicative of a feature of a set of computer instructions;

train the machine learning model using the feature-related data from the dataset;

determine, using the machine learning model, a function of the set of computer instructions based on the feature-related data;

generate, by the machine learning model, a synopsis comprising a description of the function of the set of computer instructions;

receive feedback from a user indicative of whether the synopsis accurately describes the function of the set of computer instructions; and

in response to receiving the feedback, train the machine learning model with the feedback from the user.

16 . The system of claim 15 , wherein the memory stores further instructions that are further configured to cause the system to:

determine whether the feedback indicates the synopsis accurately describes the function of the set of computer instructions; and

in response to determining that the feedback indicates that the synopsis accurately describes the function of the set of computer instructions, store the synopsis in the memory.

17 . The system of claim 16 , wherein the memory stores further instructions that are further configured to cause the system to:

in response to determining the feedback indicates that the synopsis inaccurately describes the function of the set of computer instructions, generate a revised synopsis of the function of the set of computer instructions.

18 . The system of claim 17 , wherein generating a revised synopsis of the function of the set of computer instructions further comprises:

generating the revised synopsis based on the feedback, the feedback comprising edits to the synopsis made by the user.

19 . The system of claim 15 , wherein the memory stores further instructions that are further configured to cause the system to:

store the synopsis in a memory;

receive a request to search the synopsis based on a search term;

determine, based on the synopsis and the search term, whether the set of computer instructions performs a function related to the search term; and

in response to determining that the set of computer instructions performs the function related to the search term, transmit the synopsis to a user interface.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 4, 2024
From: BURGESS, BRIAN; GLANCY, BENJAMIN CABELL; PASHIKANT, NARENDER; SELLAMUTHU, GUGANATHAN
To: CAPITAL ONE SERVICES, LLC
Reel/Frame 069121/0882 →
Continuity (2)
Continuation 17515746 · Nov 1, 2021
Related Publication 20240220238A1 · Jul 4, 2024
References Cited (17)
US 8102406B2 · Peleg et al. · 2012 [cited by applicant]
US 8527979B2 · Wookey · 2013 [cited by applicant]
US 8635204B1 · Xie et al. · 2014 [cited by applicant]
US 9003366B2 · Zeidman · 2015 [cited by applicant]
US 10909313B2 · Can et al. · 2021 [cited by applicant]
US 10977258B1 · Liu et al. · 2021 [cited by applicant]
US 11392840B2 · Santhanam et al. · 2022 [cited by applicant]
US 11699432B2 · Bender · 2023 [cited by examiner]
US 11941393B2 · Burgess · 2024 [cited by examiner]
US 20170024452A1 · Budalakoti et al. · 2017 [cited by applicant]
US 20210149886A1 · Zheng et al. · 2021 [cited by applicant]
LeClair et al., “Adapting Neural Text Classification for Improved Software Categorization” 2018, IEEE, pp. 461-472. (Year: 2018). [cited by examiner]
LeClair et al., “Recommendations for Datasets for Source Code Summarization”, 2019, arXiv, 7 pages. (Year: 2019). [cited by examiner]
LeClair et al., “Improved Code Summarization via a Graph Neural Network”, 2020, ACM, pp. 184-195. (Year: 2020). [cited by examiner]
White et al., “Toward Deep Learning Software Repositories,” 2015, IEEE, pp. 334-345 (Year: 2015). [cited by applicant]
Yu et al., “CS*: Approximate Query Processing on Big Data using Scalable Join Correlated Sample Synopsis,” 2019, IEEE, pp. 583-592 (Year: 2019). [cited by applicant]
Zhang et al, “Retrieval-based Neural Source Code Summarization,” 2020, IEEE, pp. 1385-1397. (Year: 2020). [cited by applicant]