Patent Issued for GUI for configuring machine-learning services (USPTO 11392855): State Farm Mutual Automobile Insurance Company – Insurance News Net

Insurance Daily News

2022 AUG 04 (NewsRx) — By a News Reporter-Staff News Editor at Insurance Daily News — A patent by the inventors Dorner, Theodore Edward (Sugar Hill, GA, US), Murakonda, Sambasiva R. (Cumming, GA, US), filed on May 3, 2019, was published online on July 19, 2022, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents.

Patent number 11392855 is assigned to State Farm Mutual Automobile Insurance Company (Bloomington, Illinois, United States).

The following quote was obtained by the news editors from the background information supplied by the inventors: “Many machine-learning (ML) models and algorithms require large amounts of data from different sources to be developed, trained, tested, or deployed. This data may be stored in multiple sources according to multiple formats, languages, and configurations, and may be accessible via numerous different protocols and APIs. Often when developing a model, more time is spentidentifying, acquiring, sorting, or filtering the data than in all other tasks combined. Programmers and designers may spend a significant amount of time identifying data sources and writing API requests to input data, finding the appropriate data in the files sent, translating the data into a usable computer language, and storing the data appropriately for developing the model.”

In addition to the background information obtained for this patent, NewsRx journalists also obtained the inventors’ summary information for this patent: “Note, this summary has been provided to introduce a selection of concepts further described below in the detailed description. As explained in the detailed description, certain embodiments may include features and advantages not described in this summary, and certain embodiments may omit one or more features and/or advantages described in this summary.

“Described systems and techniques relate to a data pipeline tool that provides a ML design interface that a user may utilize (e.g., via an electronic device such as a personal computer, tablet, or smart phone) to design or configure data pipelines or workflows defining the manner in which ML models are developed, trained, tested, validated, or deployed. Once deployed, the designed ML model may generate predictive results based on input data fed to the ML model. The tool may present the predictive results via a GUI, and may enable the user to mark-up or otherwise interact with those predictive results. For example, the tool may enable the user to share the results (which may include a mark-up or annotation provided by a user). The tool may transmit or otherwise make available the shared results for other devices to access.”

The claims supplied by the inventors are:

“1. A system for configuring machine-learning services, the system comprising: a display; one or more processors coupled to the display; one or more memories, coupled to the one or more processors, storing: (1) a plurality of preconfigured data templates, each template of the plurality of preconfigured data templates having a respective association requirement; (2) a set of instructions that, when executed, causes the one or more processors to: select a template from the plurality of preconfigured data templates, the template having an association requirement and a set of template configurations; identify at least one of a data source, a data destination, or a data manipulation machine learning (ML) model that satisfies the association requirement of the template; display a graphical user interface (GUI) including a canvas area configured to receive input from the user, the canvas area including one or more selectable icons representing the at least one of the data source, the data destination, or the data manipulation ML model; receive a first input from the user via the one or more selectable icons, the first input indicating a selection of a first one of the at least one of the data source, the data destination, or the data manipulation ML model; receive a second input from the user, via the canvas area and corresponding to the selection, indicating a permissible use of the first one of the at least one of the data source, the data destination, or the data manipulation ML model; modify the set of template configurations, based at least in part on the first input and the second input, to create a set of modified template configurations, wherein modifying the set of template configurations includes generating an association between the template and the first one of the at least one of the data source, the data destination, or the data manipulation ML model, the set of modified template configurations including the association and being formatted in accordance with the permissible use; receive a third input from the user via the GUI; instantiate a workflow object based on the third input, the workflow object including the template and the set of modified template configurations wherein the third input comprises natural language audio input received via a tool associated with the GUI; and implement the workflow object in accordance with the set of modified template configurations, wherein implementing the workflow object includes: retrieving data from a first data source that satisfies the association requirement; executing a first data manipulation ML model that satisfies the association requirement using the retrieved data, wherein executing the first data manipulation ML model includes training the first data manipulation ML model based on objects displayed in the GUI that are dragged by the user onto the canvas area, generating execution results wherein the execution results indicate one of estimated repair costs for damage to property, a severity of a weather event and an underwriting decision for an insurance policy, storing the execution results in a first data destination that satisfies the association requirement, displaying, via the GUI, execution results generated by the workflow object and the first data manipulation ML model, receiving fourth input from a user that annotates the displayed execution results, and sharing, in response to user interaction with the GUI, the annotated execution results with an external client device.

“2. The system of claim 1, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use includes a network address for the first data source.

“3. The system of claim 1, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use comprises use of data received from the data source for at least one of training data, validation data, or testing data.

“4. The system of claim 1, wherein the preconfigured data templates include: an application programming interface (API) for a datastore type represented by the first data source.

“5. The system of claim 1, wherein the set of instructions further causes the one or more processors to receive a fourth input from the user that causes a placement in the canvas area of a link graphic connecting a first icon representing the first data source to a second icon representing the first data manipulation ML model template.

“6. The system of claim 1, wherein the set of instructions further causes the one or more processors to: (i) detect a placement in the canvas area of a first icon, selected from the one or more selectable icons, representing a second data destination; (ii) detect a fourth input from the user via the GUI that: (a) modifies the set of template configurations to create a second set of template configurations; and (b) indicates that the second data destination is a data destination for predictive results generated by the first data manipulation ML model; and (iii) instantiate a second workflow object such that the second workflow object inherits the second set of template configurations.

“7. A method for configuring machine-learning services, the method comprising: storing, by one or more memories, a plurality of preconfigured data templates, each template of the plurality of preconfigured data templates having a respective association requirement; selecting, via one or more processors, a template from the plurality of preconfigured data templates, the template having an association requirement and a set of template configurations; identifying, via the one or more processors, at least one of a data source, a data destination, or a data manipulation machine learning (ML) model that satisfies the association requirement of the template; displaying, via the one or more processors, a graphical user interface (GUI) including a canvas area configured to receive input from the user, the canvas area including one or more selectable icons representing the at least one of a data source, the data destination, or the data manipulation machine learning (ML) model; receiving, via the one or more processors, a first input from the user via the one or more selectable icons, the first input indicating a selection of a first one of the at least one of the data source, the data destination, or the data manipulation ML model; receiving, via the one or more processors, a second input from the user, via the canvas area and corresponding to the selection, indicating a permissible use of the first one of the at least one of the data source, the data destination, or the data manipulation ML model; modifying, via the one or more processors, the set of template configurations, based at least in part on the first input and the second input, to create a set of modified template configurations, wherein modifying the set of template configurations includes generating an association between the template and the first one of the at least one of the data source, the data destination, or the data manipulation ML model, the set of modified template configurations including the association and being formatted in accordance with the permissible use, receiving, via the one or more processors, a third input from the user via the GUI; instantiating, via the one or more processors, a workflow object based on the third input, the workflow object including the template and the set of modified template configurations wherein the third input comprises natural language audio input received via a tool associated with the GUI; and implementing the workflow object in accordance with the set of modified template configurations, wherein implementing the workflow object includes: retrieving data from a first data source that satisfies the association requirement; executing a first data manipulation ML model that satisfies the association requirement using the retrieved data, wherein executing the first data manipulation ML model includes: training the first data manipulation ML model based on objects displayed in the GUI that are dragged by the user onto the canvas area, generating execution results wherein the execution results indicate one of estimated repair costs for damage to property, a severity of a weather event and an underwriting decision for an insurance policy, storing the execution results in a first data destination that satisfies the association requirement, displaying, via the GUI, execution results generated by the workflow object and the first data manipulation ML model, receiving fourth input from a user that annotates the displayed execution results, and sharing, in response to user interaction with the GUI, the annotated execution results with an external client device.

“8. The method of claim 7, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use includes a network address for the first data source.

“9. The method of claim 7, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use comprises use of data received from the data source for at least one of training data, validation data, or testing data.

“10. The method of claim 7, wherein the preconfigured data templates include: an application programming interface (API) for a datastore type represented by the first data source.

“11. The method of claim 7, the method further comprising receiving a fourth input from that causes a placement in the canvas area of a link graphic connecting a first icon representing the first data source to a second icon representing the first data manipulation ML model template.”

There are additional claims. Please visit full patent to read further.

URL and more information on this patent, see: Dorner, Theodore Edward. GUI for configuring machine-learning services. U.S. Patent Number 11392855, filed May 3, 2019, and published online on July 19, 2022. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11392855.PN.&OS=PN/11392855RS=PN/11392855

(Our reports deliver fact-based news of research and discoveries from around the world.)

Spread the love

Leave a Reply

Your email address will not be published.

Patent Issued for GUI for configuring machine-learning services (USPTO 11392855): State Farm Mutual Automobile Insurance Company – Insurance News Net

Insurance Daily News

2022 AUG 04 (NewsRx) — By a News Reporter-Staff News Editor at Insurance Daily News — A patent by the inventors Dorner, Theodore Edward (Sugar Hill, GA, US), Murakonda, Sambasiva R. (Cumming, GA, US), filed on May 3, 2019, was published online on July 19, 2022, according to news reporting originating from Alexandria, Virginia, by NewsRx correspondents.

Patent number 11392855 is assigned to State Farm Mutual Automobile Insurance Company (Bloomington, Illinois, United States).

The following quote was obtained by the news editors from the background information supplied by the inventors: “Many machine-learning (ML) models and algorithms require large amounts of data from different sources to be developed, trained, tested, or deployed. This data may be stored in multiple sources according to multiple formats, languages, and configurations, and may be accessible via numerous different protocols and APIs. Often when developing a model, more time is spentidentifying, acquiring, sorting, or filtering the data than in all other tasks combined. Programmers and designers may spend a significant amount of time identifying data sources and writing API requests to input data, finding the appropriate data in the files sent, translating the data into a usable computer language, and storing the data appropriately for developing the model.”

In addition to the background information obtained for this patent, NewsRx journalists also obtained the inventors’ summary information for this patent: “Note, this summary has been provided to introduce a selection of concepts further described below in the detailed description. As explained in the detailed description, certain embodiments may include features and advantages not described in this summary, and certain embodiments may omit one or more features and/or advantages described in this summary.

“Described systems and techniques relate to a data pipeline tool that provides a ML design interface that a user may utilize (e.g., via an electronic device such as a personal computer, tablet, or smart phone) to design or configure data pipelines or workflows defining the manner in which ML models are developed, trained, tested, validated, or deployed. Once deployed, the designed ML model may generate predictive results based on input data fed to the ML model. The tool may present the predictive results via a GUI, and may enable the user to mark-up or otherwise interact with those predictive results. For example, the tool may enable the user to share the results (which may include a mark-up or annotation provided by a user). The tool may transmit or otherwise make available the shared results for other devices to access.”

The claims supplied by the inventors are:

“1. A system for configuring machine-learning services, the system comprising: a display; one or more processors coupled to the display; one or more memories, coupled to the one or more processors, storing: (1) a plurality of preconfigured data templates, each template of the plurality of preconfigured data templates having a respective association requirement; (2) a set of instructions that, when executed, causes the one or more processors to: select a template from the plurality of preconfigured data templates, the template having an association requirement and a set of template configurations; identify at least one of a data source, a data destination, or a data manipulation machine learning (ML) model that satisfies the association requirement of the template; display a graphical user interface (GUI) including a canvas area configured to receive input from the user, the canvas area including one or more selectable icons representing the at least one of the data source, the data destination, or the data manipulation ML model; receive a first input from the user via the one or more selectable icons, the first input indicating a selection of a first one of the at least one of the data source, the data destination, or the data manipulation ML model; receive a second input from the user, via the canvas area and corresponding to the selection, indicating a permissible use of the first one of the at least one of the data source, the data destination, or the data manipulation ML model; modify the set of template configurations, based at least in part on the first input and the second input, to create a set of modified template configurations, wherein modifying the set of template configurations includes generating an association between the template and the first one of the at least one of the data source, the data destination, or the data manipulation ML model, the set of modified template configurations including the association and being formatted in accordance with the permissible use; receive a third input from the user via the GUI; instantiate a workflow object based on the third input, the workflow object including the template and the set of modified template configurations wherein the third input comprises natural language audio input received via a tool associated with the GUI; and implement the workflow object in accordance with the set of modified template configurations, wherein implementing the workflow object includes: retrieving data from a first data source that satisfies the association requirement; executing a first data manipulation ML model that satisfies the association requirement using the retrieved data, wherein executing the first data manipulation ML model includes training the first data manipulation ML model based on objects displayed in the GUI that are dragged by the user onto the canvas area, generating execution results wherein the execution results indicate one of estimated repair costs for damage to property, a severity of a weather event and an underwriting decision for an insurance policy, storing the execution results in a first data destination that satisfies the association requirement, displaying, via the GUI, execution results generated by the workflow object and the first data manipulation ML model, receiving fourth input from a user that annotates the displayed execution results, and sharing, in response to user interaction with the GUI, the annotated execution results with an external client device.

“2. The system of claim 1, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use includes a network address for the first data source.

“3. The system of claim 1, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use comprises use of data received from the data source for at least one of training data, validation data, or testing data.

“4. The system of claim 1, wherein the preconfigured data templates include: an application programming interface (API) for a datastore type represented by the first data source.

“5. The system of claim 1, wherein the set of instructions further causes the one or more processors to receive a fourth input from the user that causes a placement in the canvas area of a link graphic connecting a first icon representing the first data source to a second icon representing the first data manipulation ML model template.

“6. The system of claim 1, wherein the set of instructions further causes the one or more processors to: (i) detect a placement in the canvas area of a first icon, selected from the one or more selectable icons, representing a second data destination; (ii) detect a fourth input from the user via the GUI that: (a) modifies the set of template configurations to create a second set of template configurations; and (b) indicates that the second data destination is a data destination for predictive results generated by the first data manipulation ML model; and (iii) instantiate a second workflow object such that the second workflow object inherits the second set of template configurations.

“7. A method for configuring machine-learning services, the method comprising: storing, by one or more memories, a plurality of preconfigured data templates, each template of the plurality of preconfigured data templates having a respective association requirement; selecting, via one or more processors, a template from the plurality of preconfigured data templates, the template having an association requirement and a set of template configurations; identifying, via the one or more processors, at least one of a data source, a data destination, or a data manipulation machine learning (ML) model that satisfies the association requirement of the template; displaying, via the one or more processors, a graphical user interface (GUI) including a canvas area configured to receive input from the user, the canvas area including one or more selectable icons representing the at least one of a data source, the data destination, or the data manipulation machine learning (ML) model; receiving, via the one or more processors, a first input from the user via the one or more selectable icons, the first input indicating a selection of a first one of the at least one of the data source, the data destination, or the data manipulation ML model; receiving, via the one or more processors, a second input from the user, via the canvas area and corresponding to the selection, indicating a permissible use of the first one of the at least one of the data source, the data destination, or the data manipulation ML model; modifying, via the one or more processors, the set of template configurations, based at least in part on the first input and the second input, to create a set of modified template configurations, wherein modifying the set of template configurations includes generating an association between the template and the first one of the at least one of the data source, the data destination, or the data manipulation ML model, the set of modified template configurations including the association and being formatted in accordance with the permissible use, receiving, via the one or more processors, a third input from the user via the GUI; instantiating, via the one or more processors, a workflow object based on the third input, the workflow object including the template and the set of modified template configurations wherein the third input comprises natural language audio input received via a tool associated with the GUI; and implementing the workflow object in accordance with the set of modified template configurations, wherein implementing the workflow object includes: retrieving data from a first data source that satisfies the association requirement; executing a first data manipulation ML model that satisfies the association requirement using the retrieved data, wherein executing the first data manipulation ML model includes: training the first data manipulation ML model based on objects displayed in the GUI that are dragged by the user onto the canvas area, generating execution results wherein the execution results indicate one of estimated repair costs for damage to property, a severity of a weather event and an underwriting decision for an insurance policy, storing the execution results in a first data destination that satisfies the association requirement, displaying, via the GUI, execution results generated by the workflow object and the first data manipulation ML model, receiving fourth input from a user that annotates the displayed execution results, and sharing, in response to user interaction with the GUI, the annotated execution results with an external client device.

“8. The method of claim 7, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use includes a network address for the first data source.

“9. The method of claim 7, wherein the first one of the at least one of the data source, the data destination, or the data manipulation ML model comprises the data source, and wherein the permissible use comprises use of data received from the data source for at least one of training data, validation data, or testing data.

“10. The method of claim 7, wherein the preconfigured data templates include: an application programming interface (API) for a datastore type represented by the first data source.

“11. The method of claim 7, the method further comprising receiving a fourth input from that causes a placement in the canvas area of a link graphic connecting a first icon representing the first data source to a second icon representing the first data manipulation ML model template.”

There are additional claims. Please visit full patent to read further.

URL and more information on this patent, see: Dorner, Theodore Edward. GUI for configuring machine-learning services. U.S. Patent Number 11392855, filed May 3, 2019, and published online on July 19, 2022. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f=G&l=50&s1=11392855.PN.&OS=PN/11392855RS=PN/11392855

(Our reports deliver fact-based news of research and discoveries from around the world.)

Spread the love

Leave a Reply

Your email address will not be published.