# US 11,443,239 B2 · Interface for machine teaching modeling > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,443,239 B2 - Document ID: US11443239B2 - Status: Granted - Kind: B2 - Application number: 16/821,919 - Filed: 2020-03-17 - Published: 2021-09-23 - Granted: 2022-09-13 - Earliest claimed: 2020-03-17 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Sean Squires; Kristopher William Paries; Mingquan Xue; Krishna Kant Gupta; Chunxu Li; Anamika Bedi; Qisheng Chen; Micaela Osuji; Nicholas Anthony Buelich, II - Web page: https://patentgrep.com/patent/US11443239B2 ## CPC - G06F 16/35: Clustering; Classification - G06N 20/00: Machine learning - G06F 9/451: Execution arrangements for user interfaces - G06F 40/166: Editing, e.g. inserting or deleting - G06F 16/1734: {Details of monitoring file system events, e.g. by the use of hooks, filter drivers, logs} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract Techniques configuring a machine learning model include instantiating a user interface configured to communicate with a machine learning model hosted on a collaborative computing platform. A selection of a file for input to the machine learning model, a selection of content in the file for input to the machine learning model, and instructions for applying the selected content to the machine learning model are received and sent to the machine learning model. A selection of one or more directories and an instruction to apply the machine learning model are sent to the machine learning model. ## Claims ### Claim 1 A method for configuring a machine learning model, the method comprising: instantiating a user interface configured to communicate with a machine learning model hosted on and integrated with a collaborative computing platform hosting a plurality of applications, the collaborative computing platform configured to manage execution and processing of uploaded documents for collaboration between one or more users in a distributed system, and retain and modify metadata associated with the uploaded documents; receiving, via the user interface, a selection of an uploaded document for input to the machine learning model, wherein the uploaded document is stored on the collaborative computing platform and the uploaded document is associated with one of the plurality of applications for creating and editing the document; receiving, via an interactive area on the user interface, a selection of content in the uploaded document for input to the machine learning model; receiving, via the interactive area, one or more instructions for applying the selected content to the machine learning model; communicating, by the user interface to the machine learning model, data indicative of the selected uploaded document, selected content, and instructions; rendering, on the user interface, a first option to enter additional inputs to the machine learning model and a second option to publish the machine learning model; in response to selection of the first option, receiving, by the user interface, additional inputs to the machine learning model; and in response to selection of the second option, publishing the machine learning model by: receiving an indication of an uploaded document of the collaborative computing platform to be processed by the machine learning model; and sending an instruction to the machine learning model to apply the machine learning model to the indicated uploaded document; wherein: the machine learning model is trained using the selected uploaded document, selected content, and instructions; the trained machine learning model is operable to generate outputs based on the indicated uploaded document as input data; the trained model and inputs are published using existing data structures of the collaborative computing platform independent of a separate artificial intelligence platform; the uploaded document is classified and key metadata is extracted and written directly to the existing data structures of the collaborative computing platform; and when additional documents are subsequently uploaded to the platform, the published trained model and inputs are used to classify the additional documents, extract values, and write the values to fields in the existing data structures of the collaborative computing platform. ### Claim 2 (depends on 1) The method of claim 1, wherein the machine learning model implements a machine teaching paradigm. ### Claim 3 (depends on 1) The method of claim 1, wherein the uploaded document contain text. ### Claim 4 (depends on 1) The method of claim 1, wherein the interactive area is configured to allow a user to highlight selected text as input to the machine learning model and define explanations for how to apply the selected text. ### Claim 5 (depends on 1) The method of claim 1, wherein publishing the machine learning model comprises applying the machine learning model to uploaded documents that are stored in selected directories and uploaded documents that are added to the selected directories. ### Claim 6 (depends on 1) The method of claim 1, further comprising: rendering, on the user interface, a third option to test the machine learning model; and in response to selection of the third option, sending an instruction to apply the machine learning model to selected uploaded documents. ### Claim 7 (depends on 1) The method of claim 1, further comprising rendering, on the user interface, a confidence level for the machine learning model. ### Claim 8 (depends on 1) The method of claim 1, wherein the interactive area is configured to allow a user to enter corrections to previously entered input to the machine learning model. ### Claim 9 A computing device comprising: one or more processors; a memory in communication with the one or more processors, the memory having computer-readable instructions stored thereupon which, when executed by the one or more processors, cause the computing device to: instantiate a user interface configured to communicate with a machine learning model hosted on and integrated with a collaborative computing platform hosting a plurality of applications, the collaborative computing platform configured to manage execution and processing of uploaded documents for collaboration between one or more users in a distributed system, and retain and modify metadata associated with the uploaded documents; receive, via the user interface, a selection of one or more uploaded documents for input to the machine learning model, wherein the uploaded documents are stored on the collaborative computing platform and the uploaded documents are associated with one of the plurality of applications for creating and editing the documents; receive, via an interactive area on the user interface, selections of content in the uploaded documents for input to the machine learning model; receive, via the interactive area, one or more instructions for applying the selected content to the machine learning model; communicate, to the machine learning model, data indicative of the selected uploaded documents, selected content, and instructions; in response to selection of an option to publish the machine learning model: receiving a selection of a uploaded document of the collaborative computing platform; and sending an instruction to the machine learning model to apply the machine learning model to the selected uploaded document; wherein: the machine learning model is trained using the selected uploaded documents, selected content, and instructions; the trained machine learning model is operable to generate outputs based on the selected uploaded document as input data; the trained model and inputs are published using existing data structures of the collaborative computing platform without using a separate artificial intelligence platform; the uploaded document is classified and key metadata is extracted and written directly to the existing data structures of the collaborative computing platform; and when additional documents are subsequently uploaded to the platform, the published trained model and inputs are used to classify the additional documents, extract values, and write the values to fields in the existing data structures of the collaborative computing platform. ### Claim 10 (depends on 9) The computing device of claim 9, wherein the machine learning model implements a machine teaching paradigm. ### Claim 11 (depends on 9) The computing device of claim 9, wherein the interactive area is configured to allow a user to highlight selected text as input to the machine learning model and define explanations for how to apply the selected text. ### Claim 12 (depends on 9) The computing device of claim 9, wherein publishing the machine learning model comprises applying the machine learning model to uploaded documents that are stored in selected directories and uploaded documents that are added to the selected directories. ### Claim 13 (depends on 9) The computing device of claim 9, further comprising computer-readable instructions stored thereupon which, when executed by the one or more processors, cause the computing device to: render, on the user interface, an option to test the machine learning model; and in response to selection of the option to test the machine learning model, send an instruction to apply the machine learning model to selected uploaded documents. ### Claim 14 (depends on 9) The computing device of claim 9, further comprising computer-readable instructions stored thereupon which, when executed by the one or more processors, cause the computing device to: render, on the user interface, a confidence level for the machine learning model. ### Claim 15 (depends on 9) The computing device of claim 9, wherein the interactive area is configured to allow a user to enter corrections to previously entered input to the machine learning model. ### Claim 16 One or more non-transitory computer-readable media storing computer-executable instructions that, upon execution cause one or more processors of a computing device to cause the computing device to perform operations comprising: receiving, via a user interface configured to communicate with a machine learning model hosted on and integrated with a collaborative computing platform hosting a plurality of applications, the collaborative computing platform configured to manage execution and processing of uploaded documents for collaboration between one or more users in a distributed system, and retain and modify metadata associated with the uploaded documents, a selection of one or more uploaded documents for input to the machine learning model, wherein the uploaded documents are stored on the collaborative computing platform and the uploaded documents are associated with one of the plurality of applications for creating and editing the documents; receive, via an interactive area on the user interface, selections of content in the uploaded documents for input to the machine learning model; receive, via the interactive area, one or more instructions for applying the selected content to the machine learning model; communicate, to the machine learning model, data indicative of the selected uploaded documents, selected content, and instructions; in response to selection of an option to publish the machine learning model: receiving a selection of one or more directories of the collaborative computing platform; and sending an instruction to the machine learning model to apply the machine learning model to the selected directories; wherein: the machine learning model is trained using the selected uploaded document, selected content, and instructions; the trained machine learning model is operable to generate outputs based on the indicated uploaded document as input data; the trained model and inputs are published using existing data structures of the collaborative computing platform and independently of a separate artificial intelligence platform; the uploaded document is classified and key metadata is extracted and written directly to the existing data structures of the collaborative computing platform; and when additional documents are subsequently uploaded to the platform, the published trained model and inputs are used to classify the additional documents, extract values, and write the values to fields in the existing data structures of the collaborative computing platform. ### Claim 17 (depends on 16) The non-transitory computer-readable media of claim 16 further comprising computer-executable instructions that, upon execution cause one or more processors of a computing device to cause the computing device to perform operations comprising: render, on the user interface, an option to test the machine learning model; and in response to selection of the option to test the machine learning model, send an instruction to apply the machine learning model to selected directories. ### Claim 18 (depends on 16) The non-transitory computer-readable media of claim 16, wherein the interactive area is configured to allow a user to highlight selected text as input to the machine learning model and define explanations for how to apply the selected text. ### Claim 19 (depends on 16) The non-transitory computer-readable media of claim 16, wherein publishing the machine learning model comprises applying the machine learning model to directories that are stored in the selected directories and directories that are added to the selected directories. ### Claim 20 (depends on 16) The non-transitory computer-readable media of claim 16, wherein the interactive area is configured to allow a user to enter corrections to previously entered input to the machine learning model.