View scores
A home view data structure is described. The home view data structure is made up of multiple entries. Each entry contains information identifying a home; and a quantitative value scoring the view available from the identified home, such that the contents of the data structure are usable to compare the identified homes on the basis of the views available from them.
1. A computer-implemented method, comprising:
accessing a plurality of home sale transactions each identifying a home within a distinguished geographic area and a price at which the identified home was sold;
for each of the accessed home sale transactions:
determining a geographic location of the identified home;
using an elevation at the geographic location of the identified home and elevations of one or more visual features, automatically generating a map diagram that identifies the one or more visual features visible from the identified home;
using the automatically generated map diagram, automatically determining a quantitative view score characterizing a view visible from the identified home; and
accessing attributes of the identified home;
creating a training set comprising one or more training data records, wherein the one or more training data records comprise at least some of the accessed home sale transactions and the corresponding determined quantitative view scores and accessed attributes;
training a machine learning model using the created training set;
generating an estimated value of a particular home in the distinguished geographic area by applying the trained machine learning model on a quantitative view score for the particular home and one or more attributes of the particular home;
measuring a difference between the generated estimated value of the particular home and an actual value of the particular home;
modifying the trained machine learning model using the difference between the estimated value of the particular home and the actual value of the particular home; and
applying the modified machine learning model to the quantitative view score for the particular home and the attributes of the particular home to generate a revised estimated value of the particular home.
2. The method of claim 1 , further comprising storing the trained machine learning model.
3. The method of claim 1 , further comprising causing the generated revised estimated value to be displayed together with information identifying the particular home.
4. The method of claim 1 , wherein the actual value is a selling price of the particular home.
5. The method of claim 1 , wherein the quantitative view score for the identified home scores a quantity of surrounding land area determined to be visible from the identified home.
6. The method of claim 1 , wherein the map diagram is generated using ray-tracing from the elevation at the geographic location of the identified home to the elevations of the one or more visual features.
7. The method of claim 1 , wherein the quantitative view score for the identified home is adjusted for a ground slope determined at the identified home.
8. The method of claim 1 , wherein the quantitative view score for the identified home is based on a combination of (1) a quantity of surrounding land area determined to be visible from the identified home and (2) a visibility of distinctive visual features visible from the identified home.
9. The method of claim 1 , wherein the quantitative view score for the identified home is based on a quantity of surrounding land area determined to be visible from the identified home in view of terrain types indicated at the identified home and surrounding land area.
10. The method of claim 1 , wherein the quantitative view score for the identified home is based on a quantity of surrounding land area determined to be visible from the identified home in view of indications of buildings existing at the identified home and surrounding land area.
11. The method of claim 1 , wherein the quantitative view score for the identified home scores the view available from the top of the identified home.
12. A system comprising:
at least one processor;
at least one memory coupled to the at least one processor and storing instructions that, when executed by the at least one processor, perform operations, the operations comprising:
accessing a plurality of home sale transactions each identifying a home within a distinguished geographic area and a price at which the identified home was sold;
for each of the accessed home sale transactions:
determining a geographic location of the identified home;
using an elevation at the geographic location of the identified home and elevations of one or more visual features, automatically generating a map diagram that identifies the one or more visual features visible from the identified home;
using the automatically generated map diagram, automatically determining a quantitative view score characterizing a view visible from the identified home; and
accessing attributes of the identified home;
creating a training set comprising one or more training data records, wherein the one or more training data records comprise at least some of the accessed home sale transactions and the corresponding determined quantitative view scores and accessed attributes;
training a machine learning model using the created training set;
generating an estimated value of a particular home in the distinguished geographic area by applying the trained machine learning model on a quantitative view score for the particular home and one or more attributes of the particular home;
measuring a difference between the generated estimated value of the particular home and an actual value of the particular home; and
modifying the trained machine learning model using the difference between the estimated value of the particular home and the actual value of the particular home.
13. The system of claim 12 , wherein:
the identified home is within a multiple-floor building; and
the quantitative view score for the identified home scores a view from an identified floor of the multiple-floor building that the identified home is within.
14. The system of claim 12 , wherein:
the quantitative view score for the identified home is based on a set of feature preference levels determined for a distinguished user; and
the quantitative view score for the identified home predicts the distinguished user's level of regard for the view from the identified home.
15. The system of claim 12 , wherein the quantitative view score for the particular home is at least as large as a specified minimum quantitative view score.
16. The system of claim 12 , wherein the quantitative view score for the particular home is based on a specified visual feature being visible from the particular home.
17. The system of claim 16 , wherein the at least one processor is further configured to causing the generated estimated value to be displayed together with an indication that the specified visual feature is visible from the particular home.
18. A computer-readable storage medium that does not constitute a transitory propagating signal per se, the computer-readable medium having computer instructions stored thereon which, when executed by one or more processors of a computing system, cause the computing system to perform a process, the process comprising:
accessing a plurality of home sale transactions each identifying a home within a distinguished geographic area and a price at which the identified home was sold;
for each of the accessed home sale transactions:
determining a geographic location of the identified home;
using an elevation at the geographic location of the identified home and elevations of one or more visual features, automatically generating a map diagram that identifies the one or more visual features visible from the identified home;
using the automatically generated map diagram, automatically determining a quantitative view score characterizing a view visible from the identified home; and
accessing attributes of the identified home;
creating a training set comprising one or more training data records, wherein the one or more training data records comprise at least some of the accessed home sale transactions and the corresponding determined quantitative view scores and accessed attributes;
training a machine learning model using the created training set; and
generating an estimated value of a particular home in the distinguished geographic area by applying the trained machine learning model on a quantitative view score for the particular home and one or more attributes of the particular home.
19. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home scores a quantity of surrounding land area determined to be visible from the particular home.
20. The computer-readable medium of claim 18 , wherein the map diagram is generated using ray-tracing from the elevation at the geographic location of the identified home to the elevations of the one or more visual features.
21. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home is adjusted for a ground slope determined at the particular home.
22. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home is based on a combination of (1) a quantity of surrounding land area determined to be visible from the particular home and (2) a visibility of distinctive visual features from the particular home.
23. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home is based on a quantity of surrounding land area determined to be visible from the particular home in view of terrain types indicated at the particular home and surrounding land area.
24. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home is based on a quantity of surrounding land area determined to be visible from the particular home in view of indications of buildings existing at the particular home and surrounding land area.
25. The computer-readable medium of claim 18 , wherein the quantitative view score for the particular home scores the view available from the top of the particular home.