Automatically determining a current value for a home

• US 8,140,421 B1
• Filed: 01/09/2008
• Issued: 03/20/2012
• Est. Priority Date: 01/09/2008
• Status: Active Grant

• Current Assignee: MFTB Holdco Incorporated
• Original Assignee: Zillow, Inc. (Zillow Group, Inc.)
• Inventors: Xiang, Dong, Burstein, Jonathan L., Humphries, Stanley B., Bun, Yeng, Ultis, Jonathan A.
• Primary Examiner(s): Boyce, Akiba Robinson
• Application Number:
  US11/971,758
• Time in Patent Office: 1,532 Days
• Field of Search: 705/400, 705/10, 705/36, 705/37, 705/38, 705/39, 705/41
• US Class Current: 705/36
• CPC Class Codes: F16B 11/00 Connecting constructional elements or machine parts by sticking or pressing them together, e.g. cold pressure welding non-electric welding in general B23K; methods of using adhesives independently of the form of the surfaces joined C09J5/00
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 20/10 specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
• CPC Class Codes: G06Q 20/102 Bill distribution or payments
• CPC Class Codes: G06Q 20/105 involving programming of a portable memory device, e.g. IC cards, "electronic purses"
• CPC Class Codes: G06Q 30/0278 Product appraisal
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
• CPC Class Codes: G06Q 50/16 Real estate

Abstract

A facility for valuing a distinguished home located in a distinguished geographic area is described. The facility receives home attributes for the distinguished home. For each of a plurality of valuation sub-models, the facility applies the valuation sub-model to the received home attributes to obtain a sub-model valuation for the distinguished home. The facility further applies a meta-model to the record home attributes to obtain a relative weighting factor for each sub-model. The facility then uses the obtained relative weighting factors to combine the sub-model valuations to obtain an overall valuation for the distinguished home. The facility reports the obtained valuation for the distinguished home.

Claims

1. A method in a computing system having a processor for valuing a distinguished home among a population of homes each having attributes, comprising:

retrieving information about each of a plurality of sales of homes among the population that have occurred since a distinguished date;

dividing the plurality of sales of homes among the population that have occurred since the distinguished date into a first group and a second group;

constructing a plurality of valuation sub-models, each valuation sub-model being constructed based on information about recent sales of homes among the population in the first group and attributes of the sold homes, each valuation sub-model being capable of producing a valuation for a home among the population of homes based upon attributes of the home;

constructing, by the processor a valuation meta-model based on information about recent sales of homes among the population in the first group and attributes of the sold homes, the valuation meta-model being capable of specifying, for a combination of the valuations produced for a home among the plurality of homes by each of the plurality of sub-models, the relative weight to be given to the valuation produced for the home by each of the plurality of sub-models based upon the home'"'"'s attributes;

for each of the plurality of valuation sub-models, applying the valuation sub-model to attributes of the distinguished home to obtain a valuation of the distinguished home;

applying the valuation meta-model to attributes of the distinguished home to obtain a weighting factor for each of the valuation sub-models;

combining the valuations of the distinguished home obtained from the valuation sub-models in accordance with the obtained weighting factors to obtain an overall valuation for the distinguished home; and

outputting the obtained overall valuation for the distinguished home.

2. The method of claim 1 wherein the applying and combining acts are performed for each home among the population of homes.

3. The method of claim 1 wherein at least one of the plurality of sub-models is a random forest sub-model.

4. The method of claim 1 wherein at least one of the plurality of sub-models is a K nearest neighbor sub-model.

5. The method of claim 1 wherein at least one of the plurality of sub-models is a support vector machine sub-model.

6. The method of claim 1 wherein the population of homes are those in a distinguished census tract.

7. The method of claim 1 wherein the population of homes are those in a distinguished city.

8. The method of claim 1 wherein the population of homes are those in a distinguished neighborhood.

9. The method of claim 1 wherein the population of homes are those in a distinguished subdivision.

10. The method of claim 1 wherein the population of homes are those in a distinguished zip code.

11. The method of claim 1 wherein the population of homes are those in a distinguished group of zip codes.

12. The method of claim 1 wherein the population of homes are those in a distinguished geographic region.

13. One or more computer memories collectively containing a trained home valuation meta-model state data structure for a population of homes, comprising information that specifies how to map from attributes of a home among the population of homes to a relative weight attributable to a valuation produced for the home by each of a plurality of different valuation models for homes among the plurality of homes, wherein the information comprises a recursive partitioning tree having a root nodes and a plurality of leaf nodes, the information being generated based on information about recent sales of homes among a population of homes and attributes of the sold homes, such that the contents of the data structure may be used to obtain from attributes of a home among the population of homes a relative weight attributable to a valuation produced for the home by each of a plurality of different valuation models based upon the home'"'"'s attributes, and such that the obtained relative weights may be used to combine valuations produced for the home to obtain an overall valuation for the home such that the recursive partitioning tree may be traversed from its root node to one of its leaf nodes based on attributes of the home, and wherein the obtained relative weights may be read from the traversed-to leaf node.

14. The computer memories of claim 13 wherein the memories further contain, for each of the plurality of valuation models, information that specifies how to produce a valuation of a home among the population of homes in accordance with the valuation model from attributes of the home,

such that, for each of the plurality of valuation models, the information that specifies how to produce a valuation of a home among the population of homes in accordance with the valuation model from attributes of the home may be used to produce a valuation of a home among the population of homes in accordance with the valuation model from attributes of the home.

15. One or more computer memories collectively containing a dynamic home valuation model weighting data structure, comprising a plurality of entries, each entry both specifying a different one of a plurality of valuation models for homes that are among a population of homes and specifying a relative weight to be attributed to a result produced by the specified valuation model when applied to attributes of a home to be valued among the population of homes, the relative weights specified by the entries having been dynamically determined based upon the attributes of the home to be valued using a recursive partitioning tree having a root nodes and a plurality of leaf nodes, the relative weights specified by the entries being generated based on information about recent sales of homes among the population of homes and attributes of the sold homes, such that results produced by the plurality of valuation models may be combined in accordance with the relative weights specified by the entries to obtain an overall valuation for the home to be valued.

16. A computer-readable medium containing executable instructions capable of causing a computing system to perform a method for valuing a distinguished home among a population of homes each having attributes, the method comprising:

retrieving a plurality of valuation sub-models, each valuation sub-model being based on information about recent sales of homes among the population and attributes of the sold homes, each valuation sub-model being capable of producing a valuation for a home among the population of homes based upon attributes of the home;

retrieving a valuation meta-model based on information about recent sales of homes among the population and attributes of the sold homes, the valuation meta-model being capable of specifying, for a combination of the valuations produced for a home among the plurality of homes by each of the plurality of sub-models, the relative weight to be given to the valuation produced for the home by each of the plurality of sub-models based upon the home'"'"'s attributes;

for each of the plurality of valuation sub-models, applying the valuation sub-model to attributes of the distinguished home to obtain a valuation of the distinguished home;

applying the valuation meta-model to attributes of the distinguished home to obtain a weighting factor for each of the valuation sub-models; and

combining the valuations of the distinguished home obtained from the valuation sub-models in accordance with the obtained weighting factors to obtain an overall valuation for the distinguished home.

17. The method of claim 16, further comprising, prior to applying the valuation sub-models:

identifying an attribute of the distinguished home for which a value is not available; and

applying a model that predicts the value of the identified attribute based upon the values of other attributes for the population of homes to obtain an imputed value for the identified attribute,

and wherein the valuation sub-models are applied to attributes of the distinguished home that include the imputed value of the identified attribute.