Chatbot co-pilot
AssigneeBank of America Corporation
InventorsRamakrishna R. Yannam, Ion Gerald McCusker, Prejish Thomas, Ravisha Andar
Life of this patent
Estimating the term…
Abstract
Chatbots may be integrated into a customer service workflow and assist a user before, during and after a user-agent interaction. The chatbot may assist an agent during a user-agent interaction. The chatbot may provide customized responses for a target agent or user. Customized responses may be formulated based on conversation context, account information, sentiment and diagnostic tools. Chabot responses may be customized to meet habits and patterns of a target agent or user. The chatbot may crowdsource questions to other agents or users. The chatbot may employ search engines, entity and slot extraction and heat maps and clustering analysis to generate relevant responses for the agent or user.
Claims18
An artificial intelligence (“AI”) system for reducing time a first human agent spends interacting with a human user, the AI system comprising:
- an automated chatbot comprising first machine executable instructions stored on a non-transitory memory of a computer system, the first machine executable instructions when executed by a processor on the computer system:
- captures first input generated by the human user;
- generates a first automated response based on the first input;
- captures second input generated by the first human agent;
- and
- generates a second automated response based on the second input;
- and
- an agent interface that provides software tools for the first human agent to concurrently interact directly with the human user and directly with the automated chatbot, the agent interface comprising second machine executable instructions stored on the non-transitory memory of the computer system, the second machine executable instructions when executed by the processor on the computer system generates:
- a first interface window:
- presenting the first automated response;
- and
- allowing the first human agent to override the first automated response;
- a second interface window for the first human agent to interact directly with the human user;
- a third interface window:
- presenting the second automated response;
- allowing the first human agent to move the second automated response into the second interface window;
- and
- allowing the first human agent to submit third input to the automated chatbot;
- and
- a fourth interface window:
- presenting a crowdsourced response received from a second human agent;
- and
- allowing the first human agent to move the crowdsourced response into the second interface window;
- wherein, the second machine executable instructions when executed by the processor on the computer system:
- presents the first, second, third and fourth interface windows concurrently within the agent interface;
- and
- based on the second automated response, determines a location for each of the first, second, third and fourth interface windows within the agent interface.
An Artificial Intelligence (“AI”) system for reducing remote interaction time of a first human agent and a human user, the AI system comprising:
- an automated chatbot that:
- receives first input from the human user;
- receives second input from the first human agent;
- autonomously generates first output responsive to the first input;
- and
- autonomously generates second output responsive to the second input;
- a user interface that provides access to software tools for the human user to:
- submit the first input;
- and
- interact with the automated chatbot or the first human agent;
- an agent interface that provides software tools for the first human agent to:
- interact directly with the human user;
- and
- interact directly with the automated chatbot;
- wherein, the agent interface comprises machine executable instructions that when executed by a processor on a computer system presents:
- a first interface window showing:
- first interaction of the human user and the automated chatbot;
- and
- the first output responsive to the first input;
- a second interface window showing:
- second interaction of the first human agent with the human user;
- and
- the second output responsive to the second input;
- a third interface window for the first human agent to interact directly with the human user;
- and
- a fourth interface window showing crowdsourced responses based on the first input or the second input and received from one or more second human agents;
- wherein the automated chatbot dynamically configures the agent interface to prevent the first human agent from overriding the first output when the crowdsourced response corresponds to the first output.
An artificial intelligence (“AI”) method for reducing time a human agent spends interacting with a human user, the AI method comprising:
- using a client application running on a mobile device, initiating a first interaction between the human user and an automated chatbot;
- connecting the client application to a first agent application;
- using the first agent application, in parallel initiating:
- a second interaction between the human agent and the human user;
- a third interaction that crowdsources input provided to the first interaction and the second interaction among one or more second agent applications;
- and
- a fourth interaction between the human agent and the automated chatbot;
- and
- applying one or more machine learning algorithms to the first interaction, the second interaction and the third interaction to generate automated responses for the fourth interaction;
- wherein the method further comprises dynamically preventing the first agent application from overriding the fourth interaction when the one or more second agent applications' input correspond to the fourth interaction.
Dependencies are inferred from the claim text.