Bank of America Corporation

Pravin Kumar Sankari Bhagavathiappan

Amit Mishra

Yogesh Raghuvanshi

Nipun Mahajan

Ravisha Andar

Ramakrishna R Yannam

Emad Noorizadeh

Sushil Golani

Krithika Viswanathan

Saahithi Chillara

Ion Gerald McCusker

Ashwini Patil

Yaksh Kumar Singh

Niharika Marisetti

Mohammed Zubair Mohammed Ajmal

Smruti Soumya Mishra

Jitender Singh

Jennifer Corzo

Priyeshkumar Chandrakant Patel

Chandra Chudeswaran Sankar

Bartholomew Sanjeevinathan

Rohit Anant Bansod

A system receives an input from a user, where the input indicates a request to perform one or more tasks. The system generates a plurality of utterances from the input. Each utterance indicates a different operation. The system determines a sequence of the operations to be performed in order to perform the one or more tasks. The system identifies a conversational software algorithm that is configured to process a respective utterance based on the topic of the utterance. The system communicates each utterance to the respective conversational software algorithm. The system receives responses from the conversational software algorithms. The system determines whether each response is valid. The system generates a final response based on the valid responses. The system outputs the final response.

Response generation from multi-intent input

Systems and processes are disclosed to leverage conversational AI across voice and digital channels by automating employee interactions with clients to deliver humanized experiences with speed and accuracy. Live phone conversation. Transcriptions between clients and agents are generated. AI assist provides automation based on client interactions. Prompts are provided to speak to customers. Real time updates and summaries are sent to client devices and streamed to customize agent sessions. Auto navigation of agent views and manipulations of the agent screen are optimized by AI. Instant client validations are enabled via audible and visual confirmations on client devices. Various other AI implementations for improved processing our similarly disclosed.

AI automated facilitation of support agent/client interactions in multi-modal communications

A system receives a set of input data streams from different data sources. At least one of the set of input data streams comprises a message that indicates an anomaly with respect to a topic associated with a software application. The system determines the topic, the anomaly, and a set of metadata associated with the topic from the set of input data streams. The set of metadata comprises an occurrence frequency of the anomaly in messages, a number of data sources from which messages are received, or a timeframe window within which the messages are received. The system assigns a set of weight values to the set of metadata and determines an accumulated weight value. If it is determined that the accumulated weight value is more than a threshold weight value, the system communicates an alert indicating to execute a countermeasure action that addresses the anomalous topic.

Anomaly detection from network traffic

Utterance building to convey user input to conversational agents

A system determines content indicated on a website where a software application is accessible. The system receives a set of rules comprising a first rule that indicates a first generated response is to be published on one or more first platforms. The system generates a first text response according to the first rule and based on the determined content, an anomaly associated with a software application, and a countermeasure action that addresses or resolves the anomaly. The system determines whether the first generated text response corresponds to an expected response. In response to determining that the first generated text response corresponds to the expected response, the system publishes the first generated text response on the one or more first platforms.

Dynamic response generation based on an anomaly associated with a software application

Methods for autocomplete-enabled complaint processing are provided. Methods may receive documents and extract terms from the documents. The documents may relate to previously stored complaints. Methods may map the terms to concepts. Methods may index the documents and mapped concepts within a search platform. Methods may receive and verify a request. Upon verification, methods may extract concepts from the request. Upon concept extraction, methods may generate a query. The query may include the one or more extracted concepts and/or any other suitable data from the request. The query may be transmitted to the search platform. The search platform may identify documents that match the request. The matched documents may be transmitted to a response generator. The response generator may utilize the extracted concepts and the identified documents to generate a response to the query. The response may be transmitted to the requestor.

Concept mapping for chat agent complaint processing

A system identifies data objects from a first piece of code in an original programming language and creates a knowledge graph that represents a relationship among the data objects. The system identifies a particular data object that is unique to the original programming language. The system determines a counterpart data object that is used for a particular purpose as the identified data object in a target programming language. The system creates a second piece of code in the target programming language by replacing the particular data object with the counterpart data object. The system determines if the second piece of code is configured to perform a task that the first piece of code is programmed to perform. In response to determining that the second piece of code is configured to perform the particular task, the system executes the second piece of code.

Source code conversion from an original computer programming language to a target programming language

A system identifies data objects from a first piece of code in an original programming language and creates a knowledge graph that represents a relationship among the data objects. The system identifies an expression syntax in the first piece of code and determines an interpretation of the expression syntax. The system creates a sentence that indicates an interpretation of a respective code line. The system determines a counterpart data object in a target programming language for each word in the created sentence. In response, the system creates a second piece of code in the target programming language, such that the second price of code satisfies syntax rules in the target programming language. In response to determining that the second piece of code is configured to perform a particular task that the first piece of code is configured to, the system executes the second piece of code.

Source code validation based on converting the source code to a non-programming language

An apparatus includes a memory and processor. The memory stores previous requests and corresponding previous responses. The processor determines that a user device transmitted a new voice request, converts the voice request into a first set of text, and transmits the text to an agent device. The processor applies the machine learning algorithm to the first set of text to generate suggested responses, by identifying patterns shared by the first set of text and a subset of the previous requests that are associated with the suggested responses. The processor transmits the suggested responses to the agent device. The processor then determines that the agent device transmitted voice signals responding to the new request. The processor converts these voice signals into a second set of text. The processor stores the first set of text as a previous request, and the second set of text as a corresponding previous response.

System and method for dynamically identifying and retrieving information responsive to voice requests

An apparatus includes a memory and a processor. The memory stores a machine learning algorithm configured to select between forwarding a request to an agent device and transmitting an automatically generated reply to the request. The processor receives feedback for a decision made by the algorithm, indicating whether the automatically generated reply includes the information sought by the request. If the algorithm decided to forward the request to the agent device, a reward is assigned to feedback that indicates that the reply does not include the information, while a punishment is assigned to feedback that indicates that the reply includes the information. If the algorithm decided to transmit the reply, a reward is assigned to feedback that indicates that the reply includes the information, and a punishment is assigned to feedback that indicates that the reply does not include the information. The processor updates the algorithm using the reward/punishment.

Growing community of inventors

Chennai, India

Pravin Kumar Sankari Bhagavathiappan