| CPC G06F 11/3684 (2013.01) | 20 Claims |
|
1. A method comprising:
executing, by one or more processors, a trained machine-learning model to engage in a natural language conversation with a user about a software test, wherein the natural language conversation involves the user describing attributes of the software test in a natural language format in response to one or more prompts from the trained machine-learning model, wherein the trained machine-learning model is configured to continue the natural language conversation at least until information usable to populate a crucial variable in a script template is received from the user, the crucial variable being a variable predesignated as crucial;
determining, by the one or more processors, the attributes of the software test based on the natural language conversation;
selecting, by the one or more processors, a predefined script template based on the attributes, the predefined script template being selected from among a plurality of predefined script templates, wherein the predefined script template includes the crucial variable;
populating, by the one or more processors, variable parameters of the predefined script template to create a runtime test script, wherein populating the variable parameters involves populating a parameter of the crucial variable based on the information and populating at least one other variable parameter based on the attributes, the runtime test script being configured to implement the software test as customized based on the attributes and the information; and
providing, by the one or more processors, the runtime test script to a running service, the running service being configured to interpret the runtime test script and thereby execute the software test consistent with the variable parameters.
|