Methods and systems for automated property valuation

• US 6,609,118 B1
• Filed: 06/21/1999
• Issued: 08/19/2003
• Est. Priority Date: 06/21/1999
• Status: Expired due to Fees

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Golibersuch, David Clarence, Bonissone, Piero Patrone, Khedkar, Pratap Shankar
• Primary Examiner(s): Trammell, James P.
• Assistant Examiner(s): SHERR, CRISTINA O
• Application Number:
  US09/337,284
• Time in Patent Office: 1,520 Days
• Field of Search: 705/400, 705/10, 705/30, 705/35, 705/36, 705/37, 705/38, 705/26, 705/27, 707/200, 707/100, 707/10
• US Class Current: 705/36.00R
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 30/0641 Shopping interfaces
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• 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 40/06 Asset management; Financial planning or analysis
• CPC Class Codes: G06Q 40/12 Accounting

Abstract

The present invention is a method and system for automating the process for valuing a property that produces an estimated value of a subject property, and a quality assessment of the estimated value, that is based on the fusion of multiple processes for valuing a property. In one embodiment, three processes for valuing a subject property are fused. The first process, called LOCVAL, uses the location and living area to provide an estimate of the subject property'"'"'s value. The second process, called AIGEN, is a generative artificial intelligence method that trains a fuzzy-neural network using a subset of cases from a case-base, and produces a run-time system to provide an estimate of the subject property'"'"'s value. The third process, called AICOMP, uses a case based reasoning process similar to the sales comparison approach to determine an estimate of the subject property'"'"'s value.

Claims

1. A method of utilizing a fused estimate for estimating the value of a subject property and providing a quality measure of the estimated value, said method comprising the steps of:

providing a first system capable of utilizing a first valuation methodology to provide a first estimate of the value of the subject property and a first reliability measure of the first estimate;

providing a second system capable of utilizing a second valuation methodology to provide a second estimate of the value of the subject property and a second reliability measure of the second estimate;

obtaining from said first system a first estimate of the value of the subject property;

obtaining from said first system a first reliability measure of said first estimate;

obtaining from said second system a second estimate of the value of the subjected property;

obtaining from said second system a second reliability measure of said second estimate;

automatically providing the fused estimate of the value of the subject property based on said first estimate and said second estimate; and

automatically providing a quality measure of said fused estimate based on said first reliability measure and said second reliability measure,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

2. The method of claim 1, wherein said step of automatically providing a fused estimate of the value of the subject property is based on said first estimate, said second estimate, said first reliability measure and said second reliability measure.

3. The method of claim 1, further comprising the steps of:

providing a third system capable of utilizing a third valuation methodology to provide a third estimate of the value of the subject property and a third reliability measure of the third estimate;

obtaining from said third system a third estimate of the value of the subject property;

obtaining from said third system a third reliability measure of said third estimate;

wherein said step of automatically providing the fused estimate of the value of the subject property further comprises automatically providing the fused estimate of the value of the subject property based on said first estimate, said second estimate and said third estimate; and

wherein said step of automatically providing a quality measure of said fused estimate further comprises automatically providing a quality measure of said fused estimate based on said first reliability measure, said second reliability measure and said third reliability measure.

4. The method of claim 1, wherein said first reliability measure and said second reliability measure have a value in the range of zero to one.

5. A method of utilizing a fused estimate for estimating the value of a subject property, said method comprising the steps of:

utilizing a first valuation methodology to provide a first estimate of the subject property based on multiple attributes of the subject property;

providing a first reliability measure for said first estimate;

utilizing a second valuation methodology to provide a second estimate of the subject property based on multiple comparable properties;

providing a second reliability measure for said second estimate;

estimating the value of the subject property using said first estimate, said second estimate, said first reliability measure, and said second reliability measure to produce a fused estimate of the value of the subject property;

determining the quality of the estimated value of the subject property from said first estimate, said second estimate, said first reliability measure, and said second reliability measure,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

6. The method of claim 5, further comprising the step of:

utilizing a third valuation methodology to provide a third estimate of the subject property based on the address and square footage of the subject property;

providing a third reliability measure for said third estimate;

wherein said step of providing a first estimate uses said third estimate as a basis to provide said first estimate; and

wherein said step of providing a first reliability measure uses said third reliability measure to provide said first reliability measure.

7. The method of claim 6, wherein said step of utilizing a third valuation methodology to provide a third estimate further comprises:

retrieving the values, the square footage and geographic locations of properties within a preselected geographic region;

constructing a smooth function spanning the preselected geographic region that represents a dollar per square foot value at every point of longitude and latitude in the preselected geographic region;

determining the estimated value of the subject property based on the subject property'"'"'s address and subject property'"'"'s square footage.

8. The method of claim 5, wherein said step of utilizing a first valuation methodology to provide a first estimate comprises the steps of:

providing a generative system based.on the combination of fuzzy logic systems and neural networks capable of accepting multiple inputs and providing a single output;

retrieving data for multiple properties, the data for each property including information regarding multiple attributes and selling price of the property;

training said generative system using the retrieved data, using said attribute information as inputs and adjusting said generative system to cause the output of the generative system to conform with the selling price of the property associated with the attribute information;

providing attribute information regarding the subject property to said trained generative system; and

receiving the estimated value of the subject property from said trained generative system.

9. The method of claim 5, wherein said step of utilizing a second valuation methodology to provide a second estimate comprises the steps of:

providing attribute information regarding the subject property;

retrieving comparable properties using said attribute information;

comparing attribute information associated with the retrieved comparable properties;

ordering the retrieved comparable properties in accordance with predefined preferences regarding said attribute information;

adjusting the price of the comparable properties to reflect the differences between the comparable property and the subject property;

determining the estimated value of the subject property using the adjusted price of the comparable properties.

10. A system for utilizing a fused estimate for estimating the value of a subject property and providing a quality measure of the estimated value, said system comprising:

a first system capable of utilizing a first valuation methodology to provide a first estimate of the value of the subject property and a first reliability measure of the first estimate;

a second system capable of utilizing a second valuation methodology to provide a second estimate of the value of the subject property and a second reliability measure of the second estimate;

means for processing a first estimate of the value of the subject property received from said first system and a second estimate of the value of the subject property received from said second system and providing the fused estimate of the value of the subject property based on said first estimate and said second estimate; and

for processing a first reliability measure from said first system and a second reliability measure from said second system and providing a quality measure of said fused estimate based on said first reliability measure and said second reliability measure,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

11. A system as claimed in claim 10, wherein said first system comprises:

a generative system based on the combination of fuzzy logic systems and neural networks capable of accepting multiple inputs and providing a first estimate of the subject property and a first reliability measure.

12. A system as claimed in claim 10, wherein said first system is capable of:

retrieving data for multiple properties that includes for each property information regarding multiple attributes and the selling price of the property; and

using said attribute information for each property as inputs; and

adjusting the first system to cause the output of the first system to conform to the value of the property associated with the attribute information.

13. A system as claimed in claim 10, wherein said second system is capable of:

accepting as input attribute information regarding the subject property;

retrieving comparable properties using said attribute information;

comparing the attribute information regarding the subject property with attribute information associated with each retrieved comparable property;

ordering the retrieved comparable properties in accordance with predefined preferences regarding said attribute information;

adjusting the price of each comparable property to reflect the differences between the comparable property and the subject property;

determining the estimated value of the subject property using the adjusted price of the comparable properties.

14. A method for utilizing a fused estimate for estimating the value of a subject property, comprising:

providing a first system capable of utilizing a first valuation methodology to provide a first estimate of the value of the subject property;

providing a second system capable of utilizing a second valuation methodology to provide a second estimate of the value of the subject property;

obtaining from said first system a first estimate of the value of the subject property;

obtaining from said second system a second estimate of the value of the subject property; and

automatically providing the fused estimate of the value of the subject property based on said first estimate and said second estimate,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

15. The method of claim 14, further comprising:

providing a third system capable of utilizing a third valuation methodology to provide a third estimate of the value of the subject property;

obtaining from said third system a third estimate of the value of the subject property; and

automatically providing the fused estimate of the value of the subject property based on said first estimate, said second estimate and said third estimate.

16. A method for utilizing a fused estimate for estimating the value of a subject property, comprising:

utilizing a first valuation methodology to provide a first estimate of the subject property based on multiple attributes of the subject property;

utilizing a second valuation methodology to provide a second estimate of the subject property based on multiple comparable properties; and

estimating the value of the subject property using said first estimate and said second estimate,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

17. The method of claim 16, further comprising:

utilizing a third valuation methodology to provide a third estimate of the subject property based on the address and square footage of the subject property; and

using said third estimate as a basis to provide said first estimate.

18. The method of claim 17, wherein utilizing a third valuation methodology to provide a third estimate comprises:

retrieving square footage values and addresses of properties within a preselected geographic region;

constructing a smooth function spanning the preselected geographic region that represents a dollar per square foot value at every point of longitude and latitude in the preselected geographic region; and

determining the estimated value of the subject property based on the address and square footage of the subject property.

19. The method of claim 16, wherein utilizing a first valuation methodology to provide a first estimate comprises:

providing a generative system based on the combination of fuzzy logic systems and neural networks capable of accepting multiple inputs and providing a single output;

retrieving data for multiple properties, the data for each property including information regarding multiple attributes and selling price of the property;

training said generative system with said retrieved data;

providing attribute information regarding the subject property to said trained generative system; and

receiving the estimated value of the subject property from said trained generative system.

20. The method of claim 19, wherein said training comprises using said attribute information as inputs and adjusting said generative system to cause the output of the generative system to conform with the selling price of the property associated with the attribute information.

21. The method of claim 16, wherein utilizing a second valuation methodology to provide a second estimate comprises:

providing attribute information regarding the subject property;

retrieving comparable properties using said attribute information;

comparing attribute information associated with the retrieved comparable properties;

ordering the retrieved comparable properties in accordance with predefined preferences regarding said attribute information;

adjusting the price of the comparable properties to reflect the differences between the comparable property and the subject property;

determining the estimated value of the subject property using the adjusted price of the comparable properties.

22. A system of utilizing a fused estimate for estimating the value of a subject property, comprising:

a first automated property valuation estimator capable of utilizing a first valuation methodology to provide a first estimate of the value of the subject property;

a second automated property valuation estimator capable of utilizing a second valuation methodology to provide a second estimate of the value of the subject property; and

a fusion system that processes the first estimate of the value of the subject property received from said first automated property valuation estimator and a second estimate of the value of the subject property received from said second automated property valuation estimator and provides a fused estimate of the value of the subject property based on said first estimate and said second estimate,

wherein the first valuation methodology and the second valuation methodology comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using a single valuation methodology alone.

23. A system of claim 22, wherein said first automated property valuation estimator comprises:

a generative system based on the combination of fuzzy logic systems and neural networks capable of accepting multiple inputs and providing a first estimate of the subject property.

24. A system of claim 22, wherein said first automated property valuation estimator is capable of retrieving data for multiple properties, wherein the data for each property includes a multiple of attribute information and the selling price of the property;

using said attribute information for each property as inputs; and

outputting a value that conforms to the value of the property associated with the attribute information.

25. A system of claim 22, wherein said second automated property valuation estimator is capable of accepting as input attribute information regarding the subject property;

retrieving comparable properties using said attribute information;

comparing the attribute information regarding the subject property with attribute information associated with each retrieved comparable property;

ordering the retrieved comparable properties in accordance with predefined preferences regarding said attribute information;

adjusting the price of each comparable property to reflect the differences between the comparable property and the subject property; and

determining the estimated value of the subject property using the adjusted price of the comparable properties.

26. A method for utilizing a fused estimate for estimating the value of a subject property, said method comprising the steps of:

utilizing LOCVAL to provide a first estimate of the value of the subject property;

utilizing AIGEN to provide a second estimate of the value of the subject property;

utilizing AICOMP to provide a third estimate of the value of the subject property; and

automatically providing the fused estimate of the value of the subject property based on said first estimate, said second estimate, and said third estimate;

wherein LOCVAL, AIGEN and AICOMP comprise different valuation methodologies, and wherein the fused estimate is capable of providing a better estimate of the value of the subject property than an individual estimate of the subject property produced by using either LOCVAL, AIGEN or AICOMP alone.