Dynamic data processing of structured data and adaptive contextual mining of unstructured data during skill runtime

1 . A method implemented by a computing system for processing structured and unstructured data for distinguishing between the structured and the unstructured data based on user context for facilitating dynamic selection and presentation of a filtered set of contextually relevant data to a user on a display of a user device which is selectively filtered from the structured and the unstructured data by utilizing one or more structured content processing machine-learning models and one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data, the method comprising:

identifying a user context;

identifying relevant skills that are accessible to the system to gather and process data for display within blocks of a canvas of an interface, the relevant skills being relevant to the user context;

identifying data obtained from the relevant skills;

determining whether the data is structured or unstructured;

processing the data by:

(i) selecting and applying the one or more unstructured content processing machine-learning models from a plurality of unstructured content processing machine-learning models to generate knowledge graphs for the unstructured data with knowledge graph records for portions of the unstructured data that are determined to be contextually relevant to the user context, the one or more unstructured content processing machine-learning models being selected based on a format of the unstructured data; and

(ii) selecting and applying the one or more structured content processing machine-learning models from a plurality of structured content processing machine-learning models to generate tables for the structured data with table records for portions of the structured data that are determined to be contextually relevant to the user context, the one or more structured

content processing machine-learning models being selected based on a format of the structured data;

during a user interface session in which a user is presented the interface on the user device, using the knowledge graphs and the tables to identify contextually relevant data to be displayed on the user device within the blocks of the canvas based at least in part on the user context;

generating a filtered set of contextually relevant data which is selectively filtered from the structured and the unstructured data by utilizing the one or more structured content processing machine-learning models and the one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data; and

selectively presenting the filtered set of contextually relevant data obtained from the relevant skills by presenting on the user device the filtered set of contextually relevant data identified with the knowledge graphs and the tables while refraining from presenting on the user device other portions of the data obtained from the relevant skills that are determined to be contextually less relevant than the filtered set of contextually relevant data based on the user context.

2 . The method of claim 1 , wherein the identification of the relevant skills is further based on an identified application context.

3 . The method of claim 1 , wherein the method further comprises refraining from classifying portions of the unstructured data that are determined to be contextually irrelevant to the user context and refraining from storing a knowledge graph record for the portions of the unstructured data that are determined to be contextually irrelevant to the user context.

4 . The method of claim 1 , wherein the method further comprises refraining from generating table records for portions of the structured data that are determined to be contextually irrelevant to the user context.

5 . The method of claim 1 , wherein the processing the data further comprises classifying objects within an image.

6 . The method of claim 1 , wherein the processing the data further comprises generating a transcript from an audio file.

7 . The method of claim 1 , wherein the method further includes automatically deleting an index or knowledge graph upon determining the index or knowledge graph was last accessed a predetermined period of time prior to a current period of time.

8 . A system comprising:

a processor system; and

a computer storage medium that stores computer-executable instructions that are executable by the processor system to implement a method for processing structured and unstructured data for distinguishing between the structured and the unstructured data based on user context for facilitating dynamic selection and presentation of a filtered set of contextually relevant data to a user on a display of a user device which is selectively filtered from the structured and the unstructured data by utilizing one or more structured content processing machine-learning models and one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data, and by at least causing the system to:

identify a user context;

identify relevant skills that are accessible to the system to gather and process data for display within blocks of a canvas of an interface;

identify data obtained from the relevant skills;

determine whether the data is structured or unstructured;

process the data by:

(i) selecting and applying the one or more unstructured content processing machine-learning models from a plurality of unstructured content processing machine-learning models to generate knowledge graphs for the unstructured data with knowledge graph records for portions of the unstructured data that are determined to be contextually relevant to the user context, the one or more unstructured content processing machine-learning models being selected based on a format of the unstructured data; and

(ii) selecting and applying the one or more structured content processing machine-learning models from a plurality of structured content processing machine-learning models to generate tables for the structured data with table records for portions of the structured data that are determined to be contextually relevant to the user context, the one or more structured

content processing machine-learning models being selected based on a format of the structured data;

during a user interface session in which a user is presented the interface on the user device, use the knowledge graphs and the tables to identify contextually relevant data to be displayed on the user device within the blocks of the canvas based at least in part on the user context;

generate a filtered set of contextually relevant data which is selectively filtered from the structured and the unstructured data by utilizing the one or more structured content processing machine-learning models and the one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data; and

selectively present the filtered set of contextually relevant data obtained from the relevant skills by presenting on the user device the filtered set of contextually relevant data identified with the knowledge graphs and the tables while refraining from presenting on the user device other portions of the data obtained from the relevant skills that are determined to be contextually less relevant than the filtered set of contextually relevant data based on the user context.

9 . The system of claim 8 , wherein the identification of the relevant skills is further based on an identified application context.

10 . The system of claim 8 , wherein the computer-executable instructions are further executable for causing the system to refrain from classifying portions of the unstructured data that are determined to be contextually irrelevant to the user context and refraining from storing a knowledge graph record for the portions of the unstructured data that are determined to be contextually irrelevant to the user context.

11 . The system of claim 8 , wherein the computer-executable instructions are further executable for causing the system to refrain from generating table records for

portions of the structured data that are determined to be contextually irrelevant to the user context.

12 . The system of claim 8 , wherein computer-executable instructions are further executable for causing the system to classify objects within an image.

13 . The system of claim 8 , wherein computer-executable instructions are further executable for causing the system to generate a transcript from an audio file.

14 . The system of claim 8 , wherein the computer-executable instructions are further executable for causing the system to automatically delete an index or knowledge graph upon determining the index or knowledge graph was last accessed a predetermined period of time prior to a current period of time.

15 . A hardware storage device storing computer-executable instructions that are executable by one or more hardware processors of a computing system for causing the computing system to implement a method for processing structured and unstructured data for distinguishing between the structured and the unstructured data based on user context for facilitating dynamic selection and presentation of a filtered set of contextually relevant data to a user on a display of a user device which is selectively filtered from the structured and the unstructured data by utilizing one or more structured content processing machine-learning models and one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data, and by at least causing the computing system to:

identify a user context;

identify relevant skills that are accessible to the system to gather and process data for display within blocks of a canvas of an interface;

identify data obtained from the relevant skills;

determine whether the data is structured or unstructured;

process the data by:

(i) selecting and applying the one or more unstructured content processing machine-learning models from a plurality of unstructured content processing machine-learning models to generate knowledge graphs for the unstructured data with knowledge graph records for portions of the unstructured data that are determined to be contextually relevant to the user context, the one or more unstructured content processing machine-learning models being selected based on a format of the unstructured data; and

(ii) selecting and applying the one or more structured content processing machine-learning models from a plurality of structured content processing machine-learning models to generate tables for the structured data with table records for portions of the structured data that are determined to be contextually relevant to the user context, the one or more structured

content processing machine-learning models being selected based on a format of the structured data;

during a user interface session in which a user is presented the interface on the user device, use the knowledge graphs and the tables to identify contextually relevant data to be displayed on the user device within the blocks of the canvas based at least in part on the user context;

generate a filtered set of contextually relevant data which is selectively filtered from the structured and the unstructured data by utilizing the one or more structured content processing machine-learning models and the one or more unstructured content processing machine-learning models to make the dynamic selection of the filtered set of contextually relevant data; and

selectively present the filtered set of contextually relevant data obtained from the relevant skills by presenting on the user device the filtered set of contextually relevant data identified with the knowledge graphs and the tables while refraining from presenting on the user device other portions of the data obtained from the relevant skills that are determined to be contextually less relevant than the filtered set of contextually relevant data based on the user context.

16 . The hardware storage device of claim 15 , wherein the identification of the relevant skills is further based on an identified application context.

17 . The hardware storage device of claim 15 , wherein the computer-executable instructions are further executable for causing the system to refrain from classifying portions of the unstructured data that are determined to be contextually irrelevant to the user context and refraining from storing a knowledge graph record for the portions of the unstructured data that are determined to be contextually irrelevant to the user context.

18 . The hardware storage device of claim 15 , wherein the computer-executable instructions are further executable for causing the system to refrain from generating table records for portions of the structured data that are determined to be contextually irrelevant to the user context.

19 . The hardware storage device of claim 15 , wherein computer-executable instructions are further executable for causing the system to classify objects within an image and to generate a transcript from an audio file.

20 . The hardware storage device of claim 15 , wherein the computer-executable instructions are further executable for causing the system to automatically delete an index or knowledge graph upon determining the index or knowledge graph was last accessed a predetermined period of time prior to a current period of time.