US 12,231,378 B2
Realtime conversation AI insights and deployment
Sairam Swaminathan, Bangalore (IN); and Amrendra Kumar, Bangalore (IN)
Assigned to SAP SE, Walldorf (DE)
Filed by SAP SE, Walldorf (DE)
Filed on Jun. 8, 2023, as Appl. No. 18/331,521.
Prior Publication US 2024/0414106 A1, Dec. 12, 2024
Int. Cl. H04L 51/02 (2022.01); G06F 8/658 (2018.01); G06F 40/40 (2020.01)
CPC H04L 51/02 (2013.01) [G06F 8/658 (2018.02); G06F 40/40 (2020.01)] 11 Claims
OG exemplary drawing
 
1. A computer system, comprising:
one or more processors;
one or more machine-readable medium coupled to the one or more processors and storing computer program code comprising sets of instructions executable by the one or more processors to:
operate, by a first computer system, a first version of a chatbot, the chatbot using natural language processing to conduct conversation sessions with a plurality of users, wherein the chatbot is configured to provide conversation responses to the plurality of users by triggering a particular skill from among a first plurality of skills, wherein the particular skill corresponds to an intent of a user, wherein the first plurality of skills includes a fallback skill that is triggered when no other skill of the first plurality of skills corresponds to the intent of the user;
record, by the first computer system, conversation logs of the conversation sessions of the first version of the chatbot, the conversation logs including a set of one or more conversations that triggered the fallback skill, each conversation log comprising a unique identification, a skill description of the particular skill triggered by a particular conversation, and a skill identification;
access, by a first user on a second computer system, the conversation logs through an API, the second computer system sending a payload to the first computer system comprising a unique identifier for the first user and a first date and second date, the first computer system sending to the second computer system a response payload comprising a plurality of conversation logs;
filter, by the first user on the second computer system, the conversation logs to obtain conversation logs comprising the fallback skill;
deploy, by the first user, a second version of the chatbot from the second computer system to the first computer system while the first version of the chatbot is concurrently engaged in a set of conversation sessions on the first computer system without affecting ongoing conversations with the first version of the chatbot,
wherein the second version of the chatbot comprises a second plurality of skills including the first plurality of skills, one or more new skills, and the fallback skill,
wherein the set of conversation sessions are mapped to the first version of the chatbot,
wherein the second version of the chatbot is configured to provide conversation responses to a second plurality of users using a second plurality of skills including the fallback skill,
wherein the second version of the chatbot is configured to trigger one or more new skills of the second plurality of skills instead of the fallback skill for at least a portion of the one or more conversations where the first version of the chatbot triggered the fallback skill; and
map, by the first computer system, new conversation sessions to the second version of the chatbot, the new conversation sessions occurring after the set of conversation sessions the first version of the chatbot is engaged in, wherein the set of conversation sessions the first version of the chatbot is engaged in are still mapped to the first version of the chatbot;
determining, by the first computer system, when there are no active conversations with the first version of the chatbot; and
marking, by the first computer system, the first version of the chatbot as inactive based on said determining.