US 12,664,911 B2Granted

System and method for creating a training script on a virtual laboratory

AssigneesConstructor Technology AG, Constructor Education and Research Genossenschaft

InventorsDenis Zholobov, Laurent Dedenis, Serg Bell, Stanislav Protasov

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for creating a training script on a virtual laboratory instance running in an e-learning infrastructure. The method comprises deploying a virtual laboratory instance from a template on e-learning infrastructure. The virtual laboratory instance includes preconfigured training software and an action interception tool. The method further comprises intercepting system events at the action interception tool corresponding to training software operation, pre-processing intercepted events by filtering and grouping events into an action-centric dataset, generating a training script using a machine learning model, based on an action-centric dataset, in a human-readable format, and storing the training script in a structured text-based format file.

Claims20

Dependent claims

A method for creating a training script on a virtual laboratory instance including installed training software and running in an e-learning infrastructure comprising:

on a user computing device having a processor and memory:
deploying a virtual laboratory instance as a snapshot of a system state from a template on the e-learning infrastructure, the system state comprising a plurality of system data required to recreate an instance of one or more virtual machines deployed on the e-learning infrastructure, and a plurality of training steps to perform certain events reflective of the system state, wherein the virtual laboratory instance comprises preconfigured training software being software tested on the virtual laboratory instance and an action interception tool, the virtual laboratory instance operating as a virtual machine on the user computing device, the virtual machine deployed from the template, the template corresponding to a specific state of the preconfigured training software;
intercepting system events using the action interception tool corresponding to training software operation;
pre-processing the intercepted system events by filtering all system events to obtain a subset of all system events as a group of valuable system events according to a set of rules into an action-centric dataset;
generating a training script using a machine learning model, based on the action-centric dataset, in a human-readable format, wherein the machine learning model is trained on labeled data including at least a portion of the action-centric dataset marked with corresponding text that represent test script steps;
and
storing the training script in a structured text-based format file for subsequent testing of the preconfigured training software.

4 more elements

A system to create a training script on a virtual laboratory instance including installed training software and running in an e-learning infrastructure, the system comprising:

a user computing device having a processor and memory, including:
a virtual lab instance, deployed from a lab template as a snapshot of a system state on the e-learning infrastructure, the system state comprising a plurality of system data required to recreate an instance of one or more virtual machines deployed on the e-learning infrastructure, and a plurality of training steps to perform certain events reflective of the system state, comprises a preconfigured training software being software tested on the virtual laboratory instance and an action interception tool, wherein the action interception tool is configured to intercept system events corresponding to training software operation, the virtual lab instance operating as a virtual machine on the user computing device, the virtual machine deployed from the lab template, the template corresponding to a specific state of the preconfigured training software;
a pre-processing module configured to pre-process the intercepted system events by filtering all system events to obtain a subset of all system events as a group of valuable system events according to a set of rules into an action-centric dataset;
a test script generator configured to generate a training script based on the action-centric dataset, in a human readable format, wherein the machine learning model is trained on labeled data including at least a portion of the action-centric dataset marked with corresponding text that represent test script steps;
and
a script storage configured to store the training script in a structured text-based format file for subsequent testing of the preconfigured training software.

3 more elements

A system for generating a training script, the system comprising:

a computing device having a processor and memory, including:
a virtual lab instance instantiated according to a lab template as a snapshot of a system state corresponding to a state of training software including a plurality of system data required to recreate an instance of one or more virtual machines deployed on the e-learning infrastructure, and a plurality of training steps to perform certain events reflective of the system state, the virtual lab instance providing a virtual simulation environment for the training software being software tested on the virtual lab instance and operating as a virtual machine on the computing device, the virtual machine deployed from the template, the template corresponding to a specific state of the training software;
an action interception tool configured to intercept system events corresponding to operation of the training software;
a pre-processing module configured to group the intercepted system events based on a dataset of valuable events to create a group of intercepted system events by not including a non-valuable event from one of all of the intercepted system events in the action-centric dataset according to a set of rules;
and
a test script generator configured to generate a training script based on the group of intercepted system events for subsequent testing of the training software.

3 more elements

Dependencies are inferred from the claim text.