Predictive machine learning models
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training and applying a machine learning model. One of the methods includes the actions of obtaining a plurality of data points associated with a parcel of real property; using a machine learning model to generate a prediction from the obtained plurality of data points, the prediction indicating a likelihood that the real property will satisfy a particular parameter, wherein the machine learning model is trained using a training set comprising a collection of data points associated with a labeled set of real property parcels distinct from the specified parcel of real property, the label indicating the particular parameter and corresponding value for each real property parcel of the training set; and based on the prediction, classifying the specified parcel of real property according to a determination of whether the predicted value of the parameter satisfies a threshold.
1. A method comprising:
obtaining, from one or more sources, a plurality of data points associated with a specified parcel of real property;
extracting specific mortgage information from the plurality of data points, including identifying each mortgage recorded against the parcel and an indication of whether the mortgage is open, and generating a set of potentially open mortgages;
training a machine learning model with a training set comprising a collection of data points associated with a labeled set of real property parcels distinct from the specified parcel of real property, wherein the collection of data points for each real property parcel in the training set include statistical information about the real property parcel and a retail history of the property parcel, and wherein each label of the labeled set of the real property parcels indicates whether an open mortgage and a corresponding title defect value existed;
generating a prediction for each mortgage, in the set of potentially open mortgages, from the obtained plurality of data points using the trained machine learning model, the prediction indicating a likelihood that the real property will satisfy a particular parameter including a likelihood that a potentially open mortgage attached to the specified parcel of real property is actually open;
determining whether the prediction for the particular parameter satisfies a threshold value; and
classifying the specified parcel of real property as likely having an open mortgage or not having an open mortgage based on the determination.
2. The method of claim 1 , wherein the obtained plurality of data points comprises a variety of data from a variety of data sources.
3. The method of claim 1 , further comprising:
determining whether the prediction for the particular parameter does not satisfies the threshold value, considering the mortgage closed.
4. The method of claim 1 , wherein the obtained plurality of data points associated with the specified parcel of real property used in the trained machine learning model include dates associated with a recordation of one or more mortgages and transaction data indicating dates in which ownership of the specified parcel of real property changed.
5. The method of claim 4 , further comprising:
identifying potentially open mortgages including one or more of identifying directly from parcel data or identifying indirectly from the parcel data.
6. The method of claim 5 , wherein identifying potentially open mortgages indirectly from the parcel data includes determining a presence of an unrecorded mortgage based on a recorded subordinate mortgage.
7. 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, from one or more sources, a plurality of data points associated with a specified parcel of real property;
extracting specific mortgage information from the plurality of data points, including identifying each mortgage recorded against the parcel and an indication of whether the mortgage is open, and generating a set of potentially open mortgages;
training a machine learning model with a training set comprising a collection of data points associated with a labeled set of real property parcels distinct from the specified parcel of real property, wherein the collection of data points for each real property parcel in the training set include statistical information about the real property parcel and a retail history of the property parcel, and wherein each label of the labeled set of the real property parcels indicates whether an open mortgage and a corresponding title defect value existed;
generating a prediction for each mortgage, in the set of potentially open mortgages, from the obtained plurality of data points using the trained machine learning model, the prediction indicating a likelihood that the real property will satisfy a particular parameter including a likelihood that a potentially open mortgage attached to the specified parcel of real property is actually open;
determining whether the prediction for the particular parameter satisfies a threshold value; and
classifying the specified parcel of real property as likely having an open mortgage or not having an open mortgage based on the determination.
8. The system of claim 7 , wherein the obtained plurality of data points comprises a variety of data from a variety of data sources.
9. The system of claim 7 , further comprising:
determining whether the prediction for the particular parameter does not satisfies the threshold value, considering the mortgage closed.
10. The system of claim 7 , wherein the obtained plurality of data points associated with the specified parcel of real property used in the trained machine learning model include dates associated with a recordation of one or more mortgages and transaction data indicating dates in which ownership of the specified parcel of real property changed.
11. The system of claim 10 , further comprising:
identifying potentially open mortgages including one or more of identifying directly from parcel data or identifying indirectly from the parcel data.
12. The system of claim 11 , wherein identifying potentially open mortgages indirectly from the parcel data includes determining a presence of an unrecorded mortgage based on a recorded subordinate mortgage.
13. One or more non-transitory 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, from one or more sources, a plurality of data points associated with a specified parcel of real property;
extracting specific mortgage information from the plurality of data points, including identifying each mortgage recorded against the parcel and an indication of whether the mortgage is open, and generating a set of potentially open mortgages;
training a machine learning model with a training set comprising a collection of data points associated with a labeled set of real property parcels distinct from the specified parcel of real property, wherein the collection of data points for each real property parcel in the training set include statistical information about the real property parcel and a retail history of the property parcel, and wherein each label of the labeled set of the real property parcels indicates whether an open mortgage and a corresponding title defect value existed;
generating a prediction for each mortgage, in the set of potentially open mortgages, from the obtained plurality of data points using the trained machine learning model, the prediction indicating a likelihood that the real property will satisfy a particular parameter including a likelihood that a potentially open mortgage attached to the specified parcel of real property is actually open;
determining whether the prediction for the particular parameter satisfies a threshold value; and
classifying the specified parcel of real property as likely having an open mortgage or not having an open mortgage based on the determination.
14. The one or more non-transitory computer-readable storage media of claim 13 , wherein the obtained plurality of data points comprises a variety of data from a variety of data sources.
15. The one or more non-transitory computer-readable storage media of claim 13 , further comprising:
determining whether the prediction for the particular parameter does not satisfies the threshold value, considering the mortgage closed.
16. The one or more non-transitory computer-readable storage media of claim 13 , wherein the obtained plurality of data points associated with the specified parcel of real property used in the trained machine learning model include dates associated with a recordation of one or more mortgages and transaction data indicating dates in which ownership of the specified parcel of real property changed.
17. The one or more non-transitory computer-readable storage media of claim 16 , wherein identifying potentially open mortgages indirectly from the parcel data includes determining a presence of an unrecorded mortgage based on a recorded subordinate mortgage.