Granted Patent B1

US 10,510,009 · App. 16/505,259 · Granted Dec 17, 2019

Inventor: Andy Mahdavi (San Francisco, CA)

Assignee: States Title, Inc.

G06N5/04G06N20/00G06Q40/025

Abstract

Claims (49)

1. A method comprising:

obtaining a collection of training data, the training data comprising collection of data points associated with a labeled set of real property parcels;

training a machine learning model using the training data, the machine learning model being trained to generate a likelihood with respect to a parameter from input data associated with a specific parcel of real property, wherein training includes optimizing the model using a Markov chain optimization that seeks to minimize error in the model where the model is underpinned by one or more non-differentiable functions, wherein optimizing the machine learning model includes:

selecting a first set of model parameter values;

evaluating the model performance using the initial set of model parameter values;

selecting a second set of model parameter values;

evaluating the model performance using the next set of model parameter values;

determine whether the model performance has improved;

in response to determining that the model performance improved, selecting a third set of model parameter values relative to the values of the second set of parameter values; and

in response to determining that the model performance has not improved, selecting the third set of model parameter values relative to the values of the first set of model parameter values, with some likelihood of retaining a worsened position to avoid missing a globally optimum point,

wherein one or more additional sets of parameter values are selected based on the evaluation of the model performance of the previous set until a stopping criteria is reached;

receiving a plurality of data points associated with an input parcel of real property; and

using the optimized model to generate a likelihood for the parameter for the input parcel of real property.

2. The method of claim 1 , wherein evaluating the model performance includes determining a degree of error between the model output and a known parameter value.

3. The method of claim 1 , wherein the predicted parameter is a likelihood that a mortgage attached to the specified parcel of real property is open.

4. The method of claim 1 , wherein the predicted parameter is a likelihood of a title defect affecting a parcel of real property.

5. A 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 perform operations comprising:

obtaining a collection of training data, the training data comprising collection of data points associated with a labeled set of real property parcels;

training a machine learning model using the training data, the machine learning model being trained to generate a likelihood with respect to a parameter from input data associated with a specific parcel of real property, wherein training includes optimizing the model using a Markov chain optimization that seeks to minimize error in the model where the model is underpinned by one or more non-differentiable functions, wherein optimizing the machine learning model includes:

selecting a first set of model parameter values;

evaluating the model performance using the initial set of model parameter values;

selecting a second set of model parameter values;

evaluating the model performance using the next set of model parameter values;

determine whether the model performance has improved;

in response to determining that the model performance improved, selecting a third set of model parameter values relative to the values of the second set of parameter values; and

in response to determining that the model performance has not improved, selecting the third set of model parameter values relative to the values of the first set of model parameter values, with some likelihood of retaining a worsened position to avoid missing a globally optimum point,

wherein one or more additional sets of parameter values are selected based on the evaluation of the model performance of the previous set until a stopping criteria is reached;

receiving a plurality of data points associated with an input parcel of real property; and

using the optimized model to generate a likelihood for the parameter for the input parcel of real property.

6. The system of claim 5 , wherein evaluating the model performance includes determining a degree of error between the model output and a known parameter value.

7. The system of claim 5 , wherein the predicted parameter is a likelihood that a mortgage attached to the specified parcel of real property is open.

8. The system of claim 5 , wherein the predicted parameter is a likelihood of a title defect affecting a parcel of real property.

9. One or more computer-readable storage media encoded with instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

obtaining a collection of training data, the training data comprising collection of data points associated with a labeled set of real property parcels;

training a machine learning model using the training data, the machine learning model being trained to generate a likelihood with respect to a parameter from input data associated with a specific parcel of real property, wherein training includes optimizing the model using a Markov chain optimization that seeks to minimize error in the model where the model is underpinned by one or more non-differentiable functions, wherein optimizing the machine learning model includes:

selecting a first set of model parameter values;

evaluating the model performance using the initial set of model parameter values;

selecting a second set of model parameter values;

evaluating the model performance using the next set of model parameter values;

determine whether the model performance has improved;

in response to determining that the model performance improved, selecting a third set of model parameter values relative to the values of the second set of parameter values; and

in response to determining that the model performance has not improved, selecting the third set of model parameter values relative to the values of the first set of model parameter values, with some likelihood of retaining a worsened position to avoid missing a globally optimum point,

wherein one or more additional sets of parameter values are selected based on the evaluation of the model performance of the previous set until a stopping criteria is reached;

receiving a plurality of data points associated with an input parcel of real property; and

using the optimized model to generate a likelihood for the parameter for the input parcel of real property.

10. The one or more computer-readable storage media of claim 9 , wherein evaluating the model performance includes determining a degree of error between the model output and a known parameter value.

11. The one or more computer-readable storage media of claim 9 , wherein the predicted parameter is a likelihood that a mortgage attached to the specified parcel of real property is open.

12. The one or more computer-readable storage media of claim 9 , wherein the predicted parameter is a likelihood of a title defect affecting a parcel of real property.

Assignments (9)

Cited By (2)

US 12,340,383 US 12,602,600