Signals and measurement of business goals in a chatbot platform
View Patent ↗A non-transitory computer readable medium having computer instructions stored therein that when executed by a computer system cause the computer system to perform operations including receiving a first user input defining a dialog session comprising one or more steps of a natural language conversation flow, receiving a second user input defining a positive outcome of the dialog session, receiving, via a chat interface, a natural language input, initiating the dialog session in response to an association between the natural language input and the dialog session, receiving, via the chat interface, a subsequent natural language input, and updating a data entry associated with the custom metric in an outcome log in response the subsequent natural language input being indicative of the positive outcome of the dialog session.
1 . A non-transitory computer readable medium having computer instructions stored therein that when executed by a computer system cause the computer system to perform operations comprising:
receiving a first user input defining a dialog session comprising one or more steps of a natural language conversation flow;
receiving a second user input defining a positive outcome of the dialog session;
receiving, via a chat interface, a natural language input;
initiating the dialog session in response to an association between the natural language input and the dialog session;
receiving, via the chat interface, a subsequent natural language input; and
updating a data entry associated with the custom metric in an outcome log in response the subsequent natural language input being indicative of the positive outcome of the dialog session.
2 . The non-transitory computer readable medium of claim 1 , wherein the positive outcome is identified in response to a first step in a dialog flow of the dialog session.
3 . The non-transitory computer readable medium of claim 1 , wherein the positive outcome of the dialog session is determined in response to the dialog session reaching a designated step in a dialog flow of the dialog session.
4 . The non-transitory computer readable medium of claim 1 , wherein the positive outcome of the dialog session is predefined by a chatbot administrator.
5 . The non-transitory computer readable medium of claim 1 , wherein the data entry associated with the custom metric is stored in the outcome log and wherein the outcome log includes metadata related a dialog flow of the dialog session.
6 . The non-transitory computer readable medium of claim 1 , wherein the metric is stored in an outcome log and the outcome log includes metadata related to a user.
7 . The non-transitory computer readable medium of claim 1 , wherein the positive outcome is determined in response to a database query of a database and wherein the database includes a plurality of natural language responses and wherein a natural language response is selected in response to the subsequent natural language input and displayed to a user.
8 . The non-transitory computer readable medium of claim 1 , wherein the data entry associated with the custom metric is updated in response to a user confirmation.
9 . The non-transitory computer readable medium of claim 1 , the data entry associated with the custom metric is updated in response to an absence of a subsequent user input.
10 . A method for defining a custom metric for evaluating a performance of a natural language dialog session comprising:
receiving at an administrative user interface, a first user input defining a metadata associating natural language sentence structures with structured data and a plurality of natural language response data associated with a dialog session flow;
receiving at the administrative user interface, a second user input defining an outcome data indicative of a positive outcome of the dialog session flow with at least one of a plurality of natural language responses;
initiating the dialog session;
receiving via a chat interface, a natural language query corresponding with at least one of the plurality of natural languages response per the one or more steps of the dialog session flow;
sending the natural language response to a user interface;
generating a result metadata in response natural language response; and
generating an outcome log in response to the result metadata wherein the outcome log includes an indication of the positive outcome of the dialog session.
11 . The method of claim 10 wherein the outcome log includes at least one of a subset of the result metadata, a user information and a dialog session performance metric.
12 . The method of claim 10 wherein the outcome data is associated with the result by a chatbot administrator.
13 . The method of claim 10 wherein the outcome data is associated with the natural language response and wherein the natural language response is one of a plurality of results associated with the dialog session.
14 . The method of claim 10 wherein the natural language response is associated to a result by a chatbot administrator.
15 . The method of claim 10 wherein the result metadata is generated in response to a user confirmation.
16 . The method of claim 10 wherein the result metadata is generated in response to an absence of a subsequent natural language query.
17 . The method of claim 10 wherein the dialog session is a chatbot session for performing a natural language dialog between a user application and an host application server.
18 . The method of claim 10 wherein the natural language response is one of the plurality of natural language responses associated with the dialog session and wherein the natural language response is stored in a memory communicatively coupled to an application server for performing a dialog session algorithm.
19 . A system for performing a dialog session comprising:
a memory configured for storing a plurality of natural language response data associated with a dialog session flow;
an administrator interface configured for receiving a first user input defining a metadata associating natural language sentence structures with structured data and the plurality of natural language response data associated with the dialog session flow and a second user input defining an outcome data indicative of a positive outcome of the dialog session flow with at least one of a plurality of natural language responses;
a network interface configured for receiving a natural language query, corresponding with at least one of the plurality of natural languages response per the one or more steps of the dialog session flow; and
a processor configured for sending the natural language response to a user interface, generating a result metadata in response natural language response and generating an outcome log in response to the result metadata wherein the outcome log includes an indication of the positive outcome of the dialog session.
20 . The system of claim 19 wherein the outcome log includes at least one of a subset of the results metadata, a user information and a dialog session performance metric.