BANK OF AMERICA CORPORATION patent applications published on November 30th, 2023

From WikiPatents
Revision as of 05:01, 5 December 2023 by Wikipatents (talk | contribs) (Creating a new page)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Patent applications for BANK OF AMERICA CORPORATION on November 30th, 2023

DISTRIBUTED NETWORK PROVIDING CERTIFICATE RIGHTS FOR INTELLIGENT MODELING OUTPUTS (17825141)

Main Inventor

Shailendra Singh


Brief explanation

The patent application describes a system and method for improving the authentication and verification of AI and ML model output.
  • The invention stores AI and ML decisioning data, resource data, and metadata in a non-fungible token format.
  • This data can be traced, relocated, and validated at a later time, making it more reliable and incorruptible.
  • The system ensures that the decisioning data of AI and ML model output is securely stored and can be easily verified.

Abstract

The present invention is generally related to systems and methods for providing an improved authentication and verification system for artificial intelligence (AI) and machine learning (ML) model output. The invention immutably stores AI and ML decisioning data, resource data, and metadata in a non-fungible token format such that this data can be traced, relocated, and validated at a later time. Decisioning data of AI and ML model output is incorruptible and more reliable as a result.

AGENT-SIDE CHATBOT SIMULATOR WITH THROTTLING CAPABILITIES (17751910)

Main Inventor

Ramakrishna R. Yannam


Brief explanation

- This patent application describes an application that simulates the behavior of a human agent in a chatbot testing environment.

- The application can control the amount and speed of data processed by the chatbot system to test its capabilities. - The application can determine how much data the chatbot system can handle and how quickly it can process a target volume of data. - The application can create different agent profiles for simulating conversations with varying response times. - The application does not handle chat session management to minimize resource consumption.

Abstract

This application describes apparatus and methods for an agent-side simulator for testing chatbot systems. The agent-side simulator may be configured to mimic behavior of a human agent in a chatbot testing environment. The agent-side simulator may throttle the amount of data and the frequency of speed at which the data is processed by components of the chatbot system. The agent-side simulator may test how much data that can be processed by a chatbot system and the speed which a target volume of data can be processed by the chatbot system. The agent-side simulator may build an agent profiles for different chat conversations. A first agent profile may be used to simulate conversations that require slower response times. A second agent profile may be used to simulate conversations that require faster response times. The agent-side simulator may not perform chat session handling to minimize computing resources consumed by the agent-side simulator.

PRUNING A DATABASE USING GENETIC ALGORITHM (17824015)

Main Inventor

Vijay Kumar Yarabolu


Brief explanation

The patent application describes a database manager that uses a genetic algorithm to optimize data storage and deletion in a database.
  • The database manager divides the data in the database into smaller segments and combines them randomly to form an initial generation of data chromosomes.
  • The genetic algorithm then runs multiple iterations, creating new generations of data chromosomes based on the segments with the highest optimization metrics from the previous generation.
  • The optimization metric of a segment represents its importance in the database.
  • When the genetic algorithm converges, a fitness score is calculated for each segment, indicating how many iterations it survived before the algorithm terminated.
  • Based on the fitness scores, data is deleted from the database, prioritizing segments with lower fitness scores.

Abstract

A database manager segments data stored in a database into a plurality of data segments and randomly combines the data segments into a plurality of data chromosomes that form an initial generation of a genetic algorithm. The database manager then runs multiple iterations of the genetic algorithm on the initial generation, wherein each iteration forms a new generation of data chromosomes based on data segments having the highest optimization metrics from a previous generation, wherein the optimization metric of a data segment represents a degree of importance of the data segment. When the genetic algorithm converges a fitness score is calculated for each data segment, wherein the fitness score of a data segment equals a number of iterations of the genetic algorithm the data segment survived before the genetic algorithm was terminated. Data is deleted from the database based on the fitness scores of the data segments.

SYSTEMS AND METHODS FOR VALIDATING THE ACCURACY OF AN AUTHENTICATED, END-TO-END, DIGITAL RESPONSE SYSTEM (17826240)

Main Inventor

Ganesh Chandrasekar


Brief explanation

This patent application describes a system and method for validating the accuracy of a digital response system.
  • The system curates a database of training data, including historical profile data and interaction data.
  • Profile data includes user names, identifiers, and financial instruments.
  • Interaction data includes records of user interactions with the digital response system.
  • The system uses a machine-learning engine to generate a test profile with fake information and authentication data.
  • The test profile is logged into the digital response system, and a simulated conversation is fed as input.
  • The system receives a response from the digital response system and scores its accuracy.
  • An accuracy report is generated based on the score and submitted to a system administrator.

Abstract

Systems and methods for validating the accuracy of an authenticated, end-to-end, digital response system are provided. Methods may include curating a database of training data, including historical profile data and historical interaction data. Profile data may include a name, an identifier, and a set of financial instruments for a plurality of system users. Interaction data may include records of multi-step interactions between the system users and the digital response system. Methods may include generating, via a machine-learning (ML) engine and based on the training data: a test profile including a fictitious name, a fictitious identifier, and a fictitious set of financial instruments; authentication data for the test profile including a username and password that are operational for authenticating the test profile to the digital response system; and a simulated conversation for the test profile including an utterance that is associated with an intended request. Methods may include: initiating a validation session by logging the test profile into the digital response system using the authentication data; feeding the simulated conversation as an input to the digital response system; receiving a response from the digital response system; scoring the accuracy of the response vis-à-vis the intended request; generating an accuracy report based on the accuracy score; and submitting the accuracy report to a system administrator.

ARTIFICIAL INTELLIGENCE SMART AUTOCOMPLETE (17825206)

Main Inventor

Ramakrishna R. Yannam


Brief explanation

This patent application describes a system that uses artificial intelligence to provide autocomplete suggestions within an application based on a user's search string. 
  • The AI system tracks the user's activity within one or more applications.
  • Based on this activity, the AI system creates a set of training data.
  • The AI system then trains an AI model using this training data.
  • When the user enters a search string, the AI system applies the trained model to present personalized suggestions within the application.

Abstract

This application describes apparatus and methods for providing artificial intelligence (“AI”) autocomplete that locates functionality within an application based on a search string entered by a user. The AI system may track activity of the user within one or more applications. Based on the user activity, the AI system may formulate a set of training data. The AI system may train an AI model based on the training data. After training the AI model, the AI system may detect user entry of the search string into a search function. The AI system may apply the AI model to present a target output associated with the application that is personalized for the user.

Distributed Ledger Platform for Improved Return Logistics (18231293)

Main Inventor

Maharaj Mukherjee


Brief explanation

The patent application is about return logistics and how a computing platform can facilitate the process. Here is a simplified explanation of the abstract:
  • The computing platform processes a request for transferring resources (e.g., money) from a buyer to a seller for the purchase of an item.
  • The platform records this information in a distributed ledger, which is a decentralized database.
  • Another request is received for the sale of the same item to a different buyer, and the platform processes this request as well.
  • The platform records the second transfer of resources in the distributed ledger and arranges for the item to be sent from the first buyer to the second buyer.
  • Using the distributed ledger, the platform confirms that the item has been received by the second buyer.
  • The platform then facilitates a third transfer of resources from the seller to the second buyer.

Bullet points to explain the patent/innovation:

  • The patent application introduces a computing platform that uses a distributed ledger to track and record transactions related to return logistics.
  • This platform enables the seamless transfer of resources (e.g., money) between buyers and sellers during the return process.
  • By recording all relevant information in a distributed ledger, the platform ensures transparency and accountability in the return logistics.
  • The platform also facilitates the physical movement of the item from the original buyer to the new buyer, ensuring a smooth return process.
  • The use of a distributed ledger allows for easy verification of item receipt by the new buyer, reducing the potential for disputes or fraud.
  • Overall, this innovation streamlines return logistics and enhances the efficiency and reliability of the process.

Abstract

Aspects of the disclosure relate to return logistics. A computing platform may process a first request, resulting in first information indicating a first transfer of resources from a buyer of an item to a seller. The computing platform may record, in a distributed ledger, the first information. The computing platform may receive a second request corresponding to another sale of the item, and may process the second request, which may result in second information indicating a second transfer of resources from another buyer to the seller. The computing platform may record, in the distributed ledger, the second information, and may cause the item to be sent from the buyer to the other buyer. The computing platform may identify, using the distributed ledger, that the item has been received by the other buyer, and may cause a third transfer of resources from the seller to the buyer.

Event Processing System Having Haptic Feedback (17752217)

Main Inventor

Benjamin F. Tweel


Brief explanation

The patent application describes arrangements for haptic feedback control in devices such as payment cards or mobile devices. 
  • Users can register their devices and customize haptic feedback associated with different feedback outputs.
  • When a transaction is initiated using the registered device, transaction details are analyzed to determine the appropriate feedback output.
  • A feedback output command or instruction is generated based on the identified feedback output and transmitted to the device for execution.

Abstract

Arrangements for haptic feedback control are provided. In some aspects, registration data may be received from a user. For instance, a user may identify one or more devices (e.g., payment devices such as credit or debit cards, mobile devices, or the like) associated with the user. In some examples, the user may also select customized haptic feedback associated with different feedback outputs. After receiving registration data, the computing platform may receive a request to process a transaction. The transaction may be initiated at a point-of-sale system of an entity via the payment device or card of the user. Transaction details may be analyzed to determine or identify a feedback output (e.g., approved, denied, request additional information or the like). Based on the identified feedback output, a feedback output command or instruction may be generated and transmitted to the payment device for execution.

Light-Weight and Secure Payment Processing Using a Low-Power Wide-Area Networking Protocol (17828606)

Main Inventor

Puneetha Polasa


Brief explanation

The patent application describes a method for lightweight and secure payment processing using low-power wide-area network technology. 
  • The mobile device detects communication conditions for a payment transaction.
  • The mobile device determines if low-power wide-area network communication is applicable based on the communication conditions.
  • If applicable, the mobile device identifies a payment terminal within a certain proximity.
  • The mobile device connects to the payment terminal using a low-power wide-area network protocol.
  • The mobile device sends a request to initiate the payment transaction to the payment terminal via the low-power wide-area network protocol.
  • The payment transaction is executed using the low-power wide-area network protocol.

Abstract

Aspects of the disclosure relate to light-weight and secure payment processing using low-power wide-area network technology. A mobile device may detect communication conditions associated with a payment transaction based on one or more parameters. The mobile device may identify applicability of LPWAN communication based on the communication conditions. Based on identifying the applicability of low-power wide-area network communication, the mobile device may identify a payment terminal within a threshold proximity of the mobile device. The mobile device may connect to the payment terminal via a low-power wide-area network protocol. The mobile device may transmit, via the low-power wide-area network protocol, a request to initiate the payment transaction at the payment terminal. The mobile device may cause the payment transaction to be executed using the low-power wide-area network protocol.

SMART CARD WITH INTEGRATED LOW POWER SECONDARY COMMUNICATION NETWORK (17826426)

Main Inventor

Sheetal Bhatia


Brief explanation

- This patent application describes systems and methods for creating an integrated secondary communication network.

- The method involves using a smart card to access a limited set of financial services provided by a financial institution. - The smart card can bypass the primary communication network and transmit communication packets to a gateway of the secondary communication network. - The gateway forwards the communication packets to a server of the secondary communication network. - The server then performs the requested service in response to receiving the communication packets. - The purpose of this innovation is to provide an alternative communication network for accessing financial services through a smart card.

Abstract

Systems and methods for providing an integrated secondary communication network are provided. A method may include receiving an indication of a pre-determined trigger event, and presenting, via a user interface of a smart card, an option to perform a limited subset of a set of financial services. The set of financial services may be provided to a user of the smart card by a financial institution in connection with an account to which the smart card is keyed. The method may include the smart card bypassing the primary communication network, in response to receipt of a selection to perform a service from the limited subset of the set of financial services, and transmitting, via a transmitter of the smart card to a gateway of the secondary communication network, communication packets requesting the service. The method may also include the gateway forwarding the communication packets to a server of the secondary communication network, and, in response to receipt of the communication packets, the server causing the service to be performed.

Artificial Intelligence (AI)-Based Security Systems for Monitoring and Securing Physical Locations (18232975)

Main Inventor

George Anthony Albero


Brief explanation

- The patent application is about monitoring a physical location to detect and predict anomalous activities.

- Machine learning algorithms are used to analyze inputs from various sensors, cameras, audio recording equipment, and other types of sensors. - The purpose is to detect anomalous measurements or patterns. - Notifications can be sent to devices in a network based on the detection of anomalies.

Abstract

Various aspects of the disclosure relate to monitoring a physical location to determine and/or predict anomalous activities. One or more machine learning algorithms may be used to analyze inputs from one or more sensors, cameras, audio recording equipment, and/or any other types of sensors to detect anomalous measurements/patterns. Notifications may be sent one or more devices in a network based on the detection.

System and Method for User Authentication for Information Security (17804714)

Main Inventor

Saurabh Gupta


Brief explanation

The patent application describes a system that authorizes interactions between avatars in a virtual environment by verifying the identity of the users associated with the avatars.
  • The system receives a request to authorize an interaction between a first user's avatar and a second user's avatar.
  • The system also receives a request to verify the identity of the first user.
  • An image of the first user is provided to the system, and it extracts facial features from the image.
  • The system identifies a serial number and a user credential associated with the first user.
  • Based on the facial features, serial number, and user credential, the system generates a token.
  • The system then verifies the identity of the first user by comparing the token with the provided user information.
  • If the identity of the first user is verified, the system allows the interaction between the first user's avatar and the second user's avatar.

Abstract

A system receives a request to authorize an interaction between a first avatar associated with a first user and a second avatar associated with a second user within a virtual environment. The system receives a request to verify the identity of the first user. The system receives an image of the first user and extracts facial features from the image. The system identifies a serial number and a user credential associated with the first user. The system generates a token based on the facial features, the serial number, and the user credential. The system verifies the identity of the first user based on the token. If the identity of the first user is verified, the system authorizes the interaction between the first avatar and the second avatar.

SYSTEM FOR IMPLEMENTING DYNAMIC MULTI-FACTOR SOFT LOCK ON USER IDENTIFIERS (17824360)

Main Inventor

Adithya Gadwale


Brief explanation

The abstract describes a system, computer program, and method for implementing a dynamic multi-factor soft lock on user identifiers within a distributed network.
  • The invention receives a network registration request from a user to obtain authorization for resource transfers within the network using a user identifier.
  • It queries metadata associated with the user in the network to find traceable instances of the user's digital activities.
  • Based on this metadata, it determines one or more user identifiers.
  • It then executes a soft lock on these user identifiers using an authentication subsystem.
  • After approving the authorization, it transmits control signals to display the authorization on the user's device.

Overall, the invention aims to enhance security and authentication processes within a distributed network by implementing a dynamic multi-factor soft lock on user identifiers.

Abstract

Systems, computer program products, and methods are described herein for implementing dynamic multi-factor soft lock on user identifiers. The present invention is configured to receive, from a first user input device, a network registration request from a user to obtain authorization to execute resource transfers within a distributed network using a first user identifier; query, using an information retrieval engine, metadata associated with the user within the distributed network for traceable instances of digital activities of the user; determine one or more user identifiers based on at least the metadata associated with the user; execute, using an authentication subsystem, a soft lock on the one or more user identifiers; approve authorization for the user to execute the resource transfers within the distributed network using the first user identifier; and transmit control signals configured to cause the first user device to display the authorization.

Real-Time, Intelligent Pairing and Prioritizing of Client and Server Data Queues Using Ultra-Wide Band (18192400)

Main Inventor

Shailendra SINGH


Brief explanation

The patent application describes a system that matches user devices with resource devices. Here is a simplified explanation of the abstract:
  • User preference and profile information is sent from user devices to a queue management server.
  • The user preference information is analyzed and parsed.
  • The availability of resource devices is predicted using resource information.
  • Resource devices are selected to serve user devices based on the forecast.
  • User devices are added to the end of data queues for the selected resource devices.
  • The data queues are shared between the user devices and the resource devices.

Abstract

A system is presented for matching user devices with resource devices. User preference and profile information are communicated from user devices to a queue management server. The user preference information is parsed. Availability of the resource devices is forecasted based at least in part upon the resource information and resource devices are selected to service user devices based at least in part upon the forecast. User devices are added to the end of data queues for the resource devices. The data queues are communicated to the user devices and the resource devices.

WIRELESS TRACKING AND DATA QUEUE MANAGEMENT SYSTEM (17828672)

Main Inventor

Shailendra Singh


Brief explanation

- The patent application describes a system for tracking user devices in a facility.

- User devices communicate wireless signals and user information to a wireless device and a queue management server. - The system determines the location of user devices within a defined space at different points in time. - The system tracks the movement of user devices within the defined space over a period of time using the location information. - The system identifies when user devices are within a certain distance of corresponding resource devices based on the location information. - User preference and profile information is communicated to the resource devices when user devices are within the threshold distance. - The resource devices display customized information for user devices based on their preferences and profiles.

Abstract

A system is presented for tracking user devices in a facility. Wireless signals and user information are communicated from user devices to a wireless device and a queue management server. The system determines location information for the user devices within a defined space at points in time. The system tracks the movement of the user devices within the defined space over a period based upon the location information. The system determines when user devices are within a threshold distance of corresponding resource devices based upon the determined location information. The system communicates user preference and profile information to the resource devices indicating that the user devices are within a threshold distance of the resource devices. The resource devices display customized information for user devices based upon user preference and profile information.

SYSTEM AND METHOD FOR PREVENTING GEO-LOCATION DATA TAMPERING (17825177)

Main Inventor

Shailendra Singh


Brief explanation

- The patent application describes systems and methods for preventing tampering with geo-location data.

- The method involves receiving location information for a specific location that will be used for geo-fencing purposes. - Based on the location information, one or more coordinates related to the location are identified. - A tamperproof geographic resource is created using these coordinates and stored as a non-fungible token. - The method also involves determining user features based on the tamperproof geographic resource and the location of a computing device. - These user features determine the capabilities of the computing device. - The innovation aims to ensure the integrity and accuracy of geo-location data used for geo-fencing purposes.

Abstract

Various systems, methods, and computer program products for preventing geo-location data tampering. A method of preventing geo-location data tampering is provided. The method includes receiving location information relating to a location that is to be used for geo-fencing. The method also includes identifying one or more coordinates relating to the location based on the location information. The method further includes creating a tamperproof geographic resource based on the coordinates of the location. The tamperproof geographic resource is stored as a non-fungible token. The method still further includes determining one or more user features based on the tamperproof geographic resource and a computing device location of a computing device. The one or more user features determine the capabilities of the computing device.