Patents by Inventor Chandra Khatri

Chandra Khatri has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20250259021
    Abstract: Techniques for executing system actions during conversations between a human user and an autonomous conversational system are disclosed. A first generative language model processes user messages to determine user intent, while a dialog management model analyzes the intent and conversation context to identify required system actions. The system executes actions by retrieving parameters from context, performing database queries or API calls to obtain response data, and storing results in conversation context variables. A second generative language model generates natural language responses using the action results. The system maintains conversation context including message history, action results, and state information, validates action execution, and initiates human agent handoff when needed. The system improves over time by detecting poor performance, gathering problematic conversations, and retraining using updated configurations.
    Type: Application
    Filed: April 29, 2025
    Publication date: August 14, 2025
    Inventors: Amol Kelkar, Nikhil Varghese, Chandra Khatri, Utkarsh Mittal, Nachiketa Rajpurohit, Peter Relan, Hung Tran
  • Publication number: 20250259022
    Abstract: Techniques for analyzing conversation logs to identify system actions for an autonomous conversational AI system are disclosed. Historical conversation logs are analyzed using a generative language model to identify conversation topics and subtopics. Conversations within each topic and subtopic are ranked based on frequency of occurrence and representative conversations are selected using normalized mean conversation embeddings. The selected conversations are analyzed to identify opportunities for system actions, and user messages and human agent responses are converted into system actions using a transformer-based natural language processing model. An action configuration comprising the identified system actions and required parameters is generated and stored for training the autonomous conversational AI system.
    Type: Application
    Filed: April 29, 2025
    Publication date: August 14, 2025
    Inventors: Amol Kelkar, Nikhil Varghese, Chandra Khatri, Utkarsh Mittal, Nachiketa Rajpurohit, Peter Relan, Hung Tran
  • Publication number: 20250238636
    Abstract: Techniques for analyzing conversation logs to identify system actions are described. Historical conversation logs are analyzed using a generative language model to identify conversation topics and subtopics, with conversations ranked by frequency and selected using normalized mean conversation embeddings. The system converts user messages and human agent responses into system actions using a transformer-based natural language processing model, generating database queries and API calls with required parameters. The system creates a graph structure representing conversation flows, identifies action nodes, determines required parameters, and validates the structure. An action configuration is generated and stored for training an autonomous conversational AI system. The system preprocesses logs to normalize data across channels, anonymizes personal information, and automatically improves performance by analyzing problematic conversations.
    Type: Application
    Filed: April 7, 2025
    Publication date: July 24, 2025
    Inventors: Amol Kelkar, Nikhil Varghese, Chandra Khatri, Utkarsh Mittal, Nachiketa Rajpurohit, Peter Relan, Hung Tran
  • Patent number: 12282743
    Abstract: Described herein is an Autonomous Conversational AI system, which does not require any human configuration or annotation, and is used to have multi-turn dialogs with a user. A typical Conversational AI system consists of three main models: Natural Language Understanding (NLU), Dialog Manager (DM) and Natural Language Generation (NLG), which requires human provided data and configuration. The system proposed herein leverages novel Conversational AI methods which automatically generates conversational AI configuration from any historical conversation logs. The automatically generated configuration contains Auto-Topics, Auto-Subtopics, Auto-Intents, Auto-Responses and Auto-Flows which are used to automatically train NLU, DM and NLG models. Once these models are trained for given conversation logs, the system can be used to have dialog with any user.
    Type: Grant
    Filed: January 6, 2022
    Date of Patent: April 22, 2025
    Assignee: GICRM AI LLC
    Inventors: Amol Kelkar, Nikhil Varghese, Chandra Khatri, Utkarsh Mittal, Nachiketa Rajpurohit, Peter Relan, Hung Tran
  • Publication number: 20240028658
    Abstract: One or more of the systems, apparatuses, or methods discussed herein can include a quality score for a plurality of item listings or collections of item listings. Data sparseness can be avoided, as the quality score is based on inherent properties of the listing. An item listing can be recommended to a user based on the quality score. In one or more embodiments, a method can include determining a plurality of quality scores including a quality score for each of a plurality of item listings or a plurality of collections of item listings, the quality scores determined independent of a user's attributes and independent of the user's contextual information, the contextual information corresponding to details of the user's access to a website, and recommending an item listing or collection of item listings to a user based on the quality scores and the contextual information.
    Type: Application
    Filed: October 2, 2023
    Publication date: January 25, 2024
    Applicant: eBay Inc.
    Inventors: Chandra Khatri, Steven Hui Luan, Michael Tanaka, Praveen K. Boinapalli
  • Patent number: 11809501
    Abstract: One or more of the systems, apparatuses, or methods discussed herein can include a quality score for a plurality of item listings or collections of item listings. Data sparseness can be avoided, as the quality score is based on inherent properties of the listing. An item listing can be recommended to a user based on the quality score. In one or more embodiments, a method can include determining a plurality of quality scores including a quality score for each of a plurality of item listings or a plurality of collections of item listings, the quality scores determined independent of a user's attributes and independent of the user's contextual information, the contextual information corresponding to details of the user's access to a website, and recommending an item listing or collection of item listings to a user based on the quality scores and the contextual information.
    Type: Grant
    Filed: December 30, 2014
    Date of Patent: November 7, 2023
    Assignee: eBay Inc.
    Inventors: Chandra Khatri, Steven Hui Luan, Michael Tanaka, Praveen K. Boinapalli
  • Publication number: 20230274095
    Abstract: Described herein is an Autonomous Conversational AI system, which does not require any human configuration or annotation, and is used to have multi-tum dialogs with a user. A typical Conversational AI system consists of three main models: Natural Language Understanding (NLU), Dialog Manager (DM) and Natural Language Generation (NLG), which requires human provided data and configuration. The system proposed herein leverages novel Conversational AI methods which automatically generates conversational AI configuration from any historical conversation logs. The automatically generated configuration contains Auto-Topics, Auto-Subtopics, Auto-Intents, Auto-Responses and Auto-Flows which are used to automatically train NLU, DM and NLG models. Once these models are trained for given conversation logs, the system can be used to have dialog with any user.
    Type: Application
    Filed: January 6, 2022
    Publication date: August 31, 2023
    Inventors: Amol Kelkar, Nikhil Varghese, Chandra Khatri, Utkarsh Mittal, Nachiketa Rajpurohit, Peter Relan, Hung Tran
  • Publication number: 20160063065
    Abstract: One or more of the systems, apparatuses, or methods discussed herein can include a quality score for a plurality of item listings or collections of item listings. Data sparseness can be avoided, as the quality score is based on inherent properties of the listing. An item listing can be recommended to a user based on the quality score. In one or more embodiments, a method can include determining a plurality of quality scores including a quality score for each of a plurality of item listings or a plurality of collections of item listings, the quality scores determined independent of a user's attributes and independent of the user's contextual information, the contextual information corresponding to details of the user's access to a website, and recommending an item listing or collection of item listings to a user based on the quality scores and the contextual information.
    Type: Application
    Filed: December 30, 2014
    Publication date: March 3, 2016
    Inventors: Chandra Khatri, Steven Hui Luan, Michael Tanaka, Praveen K. Boinapalli