Immersive learning experiences in application process flows
Systems and methods for providing immersive learning solutions in the context of application process flows. In one example, legacy training documents can be submitted to the system and classified and sorted based on topic and granularity. The sorted information can then be assigned to one or more learning medium buckets that will define the type of learning course experience to be generated. For example, information about a first topic at the words and sentences levels can be used to generate an interactive question-and-answer avatar that can intelligently respond to users and assist during their use of the application.
1. A method of automatically generating an interactive avatar for training human end-users in a distributed system using multiple servers over a network, the method comprising:
receiving, by at least one processor, a first document including first data via the network, wherein the at least one processor includes circuitry to facilitate operations of the multiple servers over the network;
identifying, by the at least one processor, structured electronic content, unstructured electronic content, and speech content from the first data;
generating, by the at least one processor, a first version of the first document in a common hierarchical machine readable format by executing a rule-based extraction process on the structured electronic content;
generating, by the at least one processor, a second version of the first document in the common hierarchical machine readable format by executing a machine learning model-based extraction process on the unstructured electronic content;
generating, by the at least one processor, a third version of the first document in the common hierarchical machine readable format by executing a natural language processing technique on the speech content;
generating, by the at least one processor, first hierarchical data from the common hierarchical machine readable format, based on the generated first version of the first document, the second version of the first document, and the third version of the first document;
filtering and sorting, by the at least one processor, the first data based on the first hierarchical data and four levels of granularity, including words, sentences, images, and pages, thereby producing granularized data including word data, sentence data, image data, and pages data;
assigning, by the at least one processor, content in the granularized data that includes only the word data and the sentence data to a first learning medium category;
automatically generating, by the at least one processor, a first course module based on the content assigned to the first learning medium category, wherein the first course module is generated by executing intra and multi-sentence parsing on the content in the granularized data to determine gestures, vocabulary, and speech emotions of the interactive avatar;
generating, by the at least one processor, a first interactive avatar user training experience based on the first course module; and
presenting, by the at least one processor and at a computing device, the first interactive avatar user training experience including interaction between the interactive avatar and at least one user of the human end-users.
2. The method of claim 1 , wherein the first learning medium category is one of four learning medium categories that include a presentation category, a presentation speech category, a guided navigation objects category, and an intents and responses category.
3. The method of claim 2 , further comprising:
assigning content that includes only the image data and the pages data to a second learning medium category of the four learning medium categories; and
automatically generating a second course module based on the content assigned to the second learning medium category, the second learning medium category differing from the first learning medium category.
4. The method of claim 3 , further comprising generating a second interactive avatar user training experience based on the first course module that includes a visually presented slideshow.
5. The method of claim 1 , further comprising:
recognizing a plurality of keywords in the first hierarchical data;
identifying, from the plurality of keywords, a keyword hierarchy;
creating, based on the keyword hierarchy, a training content hierarchy that includes a first topic bucket; and
sorting at least some of the granularized data to the first topic bucket based on context association.
6. The method of claim 1 , wherein the first interactive avatar user training experience includes a question-and-answer chatbot modality.
7. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to:
receive, by at least one processor, a first document including first data via a network, wherein the at least one processor includes circuitry to facilitate operations of multiple servers over the network;
identify, by the at least one processor, structured electronic content, unstructured electronic content, and speech content from the first data;
generate, by the at least one processor, a first version of the first document in a common hierarchical machine readable format by executing a rule-based extraction process on the structured electronic content;
generate, by the at least one processor, a second version of the first document in the common hierarchical machine readable format by executing a machine learning model-based extraction process the unstructured electronic content;
generate, by the at least one processor, a third version of the first document in the common hierarchical machine readable format by executing a natural language processing technique on the speech content;
generate, by the at least one processor, first hierarchical data from the common hierarchical machine readable format, based on the generated first version of the first document, the second version of the first document, and the third version of the first document;
filter and sort, by the at least one processor, the first data based on the first hierarchical data and four levels of granularity, including words, sentences, images, and pages, thereby producing granularized data including word data, sentence data, image data, and pages data;
assign, by the at least one processor, content in the granularized data that includes only the word data and the sentence data to a first learning medium category;
automatically generate, by the at least one processor, a first course module based on the content assigned to the first learning medium category, wherein the first course module is generated by executing intra and multi-sentence parsing on the content in the granularized data to determine gestures, vocabulary, and speech emotions of the interactive avatar;
generate, by the at least one processor, a first interactive avatar user training experience based on the first course module; and
present, by the at least one processor and at a computing device, the first interactive avatar user training experience including interaction between the interactive avatar and at least one user of the human end-users.
8. The non-transitory computer-readable medium storing software of claim 7 , wherein the first learning medium category is one of four learning medium categories that include a presentation category, a presentation speech category, a guided navigation objects category, and an intents and responses category.
9. The non-transitory computer-readable medium storing software of claim 7 , wherein the instructions further cause the one or more computers to:
assign content including only the image data and the pages data to a second learning medium category of the four learning medium categories; and
automatically generate a second course module based on the content assigned to the second learning medium category, the second learning medium category differing from the first learning medium category.
10. The non-transitory computer-readable medium storing software of claim 9 , wherein the instructions further cause the one or more computers to generate a second interactive avatar user training experience based on the first course module that includes a visually presented slideshow.
11. The non-transitory computer-readable medium storing software of claim 7 , wherein the instructions further cause the one or more computers to:
recognize a plurality of keywords in the first hierarchical data;
identify, from the plurality of keywords, a keyword hierarchy;
create, based on the keyword hierarchy, a training content hierarchy that includes a first topic bucket; and
sort at least some of the granularized data to the first topic bucket based on context association.
12. The non-transitory computer-readable medium storing software of claim 7 , wherein the first interactive avatar user training experience includes a question-and-answer chatbot modality.
13. A system for automatically generating an interactive avatar for training human end-users, the system comprising one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to:
receive, by at least one processor, a first document including first data via the network, wherein the at least one processor includes circuitry to facilitate operations of the multiple servers over the network;
identify, by the at least one processor, structured electronic content, unstructured electronic content, and speech content from the first data;
generate, by the at least one processor, a first version of the first document in a common hierarchical machine readable format by executing a rule-based extraction process on the structured electronic content;
generate, by the at least one processor, a second version of the first document in the common hierarchical machine readable format by executing a machine learning model-based extraction process on the unstructured electronic content;
generate, by the at least one processor, a third version of the first document in the common hierarchical machine readable format by executing a natural language processing technique on the speech content;
generate, by the at least one processor, first hierarchical data from the common hierarchical machine readable format, based on the generated first version of the first document, the second version of the first document, and the third version of the first document;
filter and sort, by the at least one processor, the first data based on the first hierarchical data and four levels of granularity, including words, sentences, images, and pages, thereby producing granularized data including word data, sentence data, image data, and pages data;
assign, by the at least one processor, content in the granularized data that includes only the word data and the sentence data to a first learning medium category;
automatically generate, by at least one processor, a first course module based on the content assigned to the first learning medium category, wherein the first course module is generated by executing intra and multi-sentence parsing on the content in the granularized data to determine gestures, vocabulary, and speech emotions of the interactive avatar;
generate, by the at least one processor, a first interactive avatar user training experience based on the first course module; and
present, by the at least one processor and at a computing device, the first interactive avatar user training experience including interaction between the interactive avatar and at least one user of the human end-users.
14. The system of claim 13 , wherein the first learning medium category is one of four learning medium categories that include a presentation category, a presentation speech category, a guided navigation objects category, and an intents and responses category.
15. The system of claim 13 , wherein the instructions further cause the one or more computers to:
assign content including only the image data and the pages data to a second learning medium category of the four learning medium categories; and
automatically generate a second course module based on the content assigned to the second learning medium category, the second learning medium category differing from the first learning medium category.
16. The system of claim 13 , wherein the instructions further cause the one or more computers to:
recognize a plurality of keywords in the first hierarchical data;
identify, from the plurality of keywords, a keyword hierarchy;
create, based on the keyword hierarchy, a training content hierarchy that includes a first topic bucket; and
sort at least some of the granularized data to the first topic bucket based on context association.
17. The system of claim 13 , wherein the first interactive avatar user training experience includes a question-and-answer chatbot modality.