Machine learning auto completion of fields

Patent metadata

Field	Value
Publication number	US-10198698-B2
Application number	US-201815939609-A
Country	US
Kind code	B2
Filing date	Mar 29, 2018
Priority date	May 4, 2017
Publication date	Feb 5, 2019
Grant date	Feb 5, 2019

Abstract

Systems and methods for using a mathematical model based on historical natural language inputs to automatically complete form fields are disclosed. An incident report may be defined with a set of required parameter fields such as category, priority, assignment, and classification. Incident report submission forms may also have other free text input fields providing information about a problem in the natural vocabulary of the person reporting the problem. Automatic completion of these so-called parameter fields may be based on analysis of the natural language inputs and use of machine learning techniques to determine appropriate values for the parameter fields. The machine learning techniques may include parsing the natural language input to determine a mathematical representation and application of the mathematical representation to “match” historically similar input. Once matched the parameter values from the historically similar input may be used instead of generic default values.

First claim

What is claimed is:

1. A method comprising:
receiving, via a cloud-based network interface from a remote client device, a request for an incident reporting form, wherein the incident reporting form includes a plurality of input fields including a first set of fields to be completed in natural language text and a second set of fields to be selected from a pre-determined set of applicable values;
obtaining, from the first set of fields, information in natural language text regarding an incident associated with the incident reporting form;
processing the information obtained from the first set of fields against a model to determine a predicted value from the pre-determined set of applicable values and a confidence level of the predicted value for at least one input field of the second set of fields of the incident reporting form;
determining that the confidence level of the predicted value is above a confidence level threshold associated with the at least one input field;
generating a confusion matrix configured to determine an accuracy of the model;
generating a cost matrix configured to determine a misclassification cost associated with the predicted value;
tuning the model to minimize the misclassification cost over a probability of an objective function;
adjusting the confidence level threshold based on the confusion matrix and the cost matrix; and
sending the predicted value to the remote client device to populate the at least one input field of the second set of fields of the incident reporting form with the predicted value.

2. The method of claim 1 , wherein the pre-determined set of applicable values are presented in a drop-down selection box where a user selects from the pre-determined set, and wherein the method further comprises:
receiving an indication of user-override of the predicted value with an alternative selection from the drop-down selection box; and
using the alternative selection to retrain the model.

3. The method of claim 1 , wherein obtaining, from the first set of fields, the information in natural language text comprises obtaining the information as the natural language text is being typed, and wherein the method further comprises:
obtaining additional information in natural language text as the additional information is input in at least one field of the first set of fields; and
dynamically readjusting the predicted value based on the additional information.

4. The method of claim 1 , further comprising:
parsing the natural language text to identify one or more N_Grams; and
processing the identified one or more N_Grams against the model to determine the predicted value from the predetermined set of applicable values and the confidence level of the predicted value.

5. The method of claim 1 , wherein the at least one input field is initially populated with a generic default value, and wherein sending the predicted value comprises sending the predicted value to the remote client device to change the generic default value to the predicted value.

6. A cloud-based computer system, comprising:
a memory partition;
a network interface communicatively coupled to one or more processing units and the memory partition, wherein the memory partition comprises computer instructions that when executed by the one or more processing units cause the cloud-based computer system to:
provide a hosted client instance communicatively coupled, via the network interface, with a remote client device;
receive a request from the remote client device for an incident reporting form, wherein the incident reporting form includes a plurality of input fields including a first set of fields to be completed in natural language text and a second set of fields to be selected from a pre-determined set of applicable values;
obtain, from the first set of fields, information in natural language text regarding an incident associated with the incident reporting form;
process the information obtained from the first set of fields against a model to determine a predicted value from the pre-determined set of applicable values and a confidence level of the predicted value for at least one input field of the second set of fields of the incident reporting form;
determine that the confidence level of the predicted value is above a confidence level threshold associated with the at least one input field;
generate a confusion matrix configured to determine an accuracy of the model;
generate a cost matrix configured to determine a misclassification cost associated with the predicted value;
tuning the model to minimize the misclassification cost over a probability of an objective function;
adjust the confidence level threshold based on the confusion matrix and the cost matrix; and
send the predicted value to the remote client device to populate the at least one input field of the second set of fields of the incident reporting form with the predicted value.

7. The cloud-based computer system of claim 6 , wherein the computer instructions further cause the cloud-based computer system to receive an indication of user-override for the predicted value and use the indication of user-override to retrain the model.

8. The cloud-based computer system of claim 6 , wherein the computer instructions further cause the cloud-based computer system to periodically retrain the model using information obtained after initial use of the model.

9. The cloud-based computer system of claim 6 , wherein the information in natural language text regarding the incident comprises a short description field generically describing the incident.

10. The cloud-based computer system of claim 6 , wherein the at least one input field value in the incident reporting form is a priority field indicative of an order in which to dedicate resources for a resolution of the incident.

11. The cloud-based computer system of claim 6 , wherein the at least one input field value in the incident reporting form is a classification field indicative of a class of a device associated with the incident.

12. The cloud-based computer system of claim 6 , wherein the at least one input field value in the incident reporting form is an assignment field indicative of a work group associated with correcting the incident.

13. The cloud-based computer system of claim 6 , wherein the computer instructions further cause the cloud-based computer system to:
parse the natural language text to identify one or more N_Grams; and
process the identified one or more N_Grams against the model to determine the predicted value from the predetermined set of applicable values and the confidence level of the predicted value for at least one input field of the second set of fields of the incident reporting form.

14. The cloud-based computer system of claim 6 , wherein the pre-determined set of applicable values is presented in a drop-down selection box where a user selects from the pre-determined set, wherein the computer instructions further cause the cloud-based computer system to:
receive an indication of user-override of the predicted value with an alternative selection from the drop-down selection box; and
use the alternative selection to retrain the model.

15. The cloud-based computer system of claim 6 , wherein the cloud-based computer system obtains, from the first set of fields, the information in natural language text regarding the incident as the natural language text is being typed, and wherein the computer instructions further cause the cloud-based computer system to:
obtain additional information in natural language text as the additional information is input in at least one fiel

Assignees

• Servicenow Inc

Inventors

• Jayaraman Baskar

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• G06F40/174

  Form filling; Merging · CPC title

• G06Q10/00

  Administration; Management · CPC title