Data mining data records to determine networks and boundaries
Systems and methods for determining a boundary from evaluation records for objects are disclosed. In some embodiments, evaluation records for objects (e.g., appraisal records for real estate properties) are accessed. The evaluation records can be used to identify objects with similar characteristics. By representing objects identified as nodes in a graph, edges between nodes representing the identified objects can be established. A network generated from the nodes and edges can be used to identify a boundary comprising objects having similar characteristics.
1. A system comprising:
non-transitory computer storage configured to store a plurality of records; and
a hardware processor in communication with the non-transitory computer storage, the hardware processor programmed to:
determine that a first object and a second object appear in a first record in the plurality of records;
determine that the first object and a third object appear in a second record in the plurality of records;
generate a network based on the determination that the first object and the second object appear in the first record and the determination that the first object and the third object appear in the second record, wherein the network identifies the first, second, and third objects as being related; and
generate a user interface displaying the network overlaying a geographic map, wherein the geographic map comprises a graphical representation of the first object, a graphical representation of the second object, and a graphical representation of the third object, wherein the geographic map comprises an indication that the first and second objects fall within a first boundary and that the third object falls within a second boundary, wherein the second boundary is nonoverlapping with, but related to, the first boundary, and wherein the second boundary is separated from the first boundary by a barrier.
2. The system of claim 1 , wherein the hardware processor is further programmed to determine that the first object is a comparable for the second object.
3. The system of claim 1 , wherein the first object comprises a real estate property.
4. The system of claim 1 , wherein the second object has a degree of separation from the first object that is less than a threshold degree of separation.
5. The system of claim 1 , wherein the hardware processor is further programmed to determine a quality status of the network, wherein the quality status of the network is based on a structure of the network.
6. The system of claim 1 , wherein the hardware processor is further programmed to append to the network geocoded information associated with the plurality of objects.
7. The system of claim 1 , wherein the first boundary comprises one of a neighborhood or a local area network.
8. The system of claim 1 , wherein the network comprises an edge between a first node representing the first object and a second node representing the second object.
9. The system of claim 1 , wherein the hardware processor is further programmed to flag the first record for inconsistency or further review in response to a determination that a comparable was not selected from within the first boundary.
10. The system of claim 9 , wherein the inconsistency comprises a fraud.
11. The system of claim 9 , wherein the first record comprises an appraisal of real estate properties.
12. The system of claim 1 , wherein the user interface displays the barrier.
13. The system of claim 1 , wherein the hardware processor is further programmed to output information associated with the first, second, and third objects via one of an electronic mail or a network-based interface.
14. A method comprising:
under control of a hardware processor:
determining that a first object and a second object appear in a first record;
determining that the first object and a third object appear in a second record;
generating a network based on the determination that the first object and the second object appear in the first record and the determination that the first object and the third object appear in the second record, wherein the network identifies the first, second, and third objects as being related; and
generating a user interface displaying the network overlaying a geographic map, wherein the geographic map comprises a graphical representation of the first object, a graphical representation of the second object, and a graphical representation of the third object, wherein the geographic map comprises an indication that the first and second objects fall within a first boundary and that the third object falls within a second boundary, wherein the second boundary is nonoverlapping with, but related to, the first boundary, and wherein the second boundary is separated from the first boundary by a barrier.
15. The method of claim 14 , wherein determining that a first object and a second object appear in a first record further comprises determining that the first object is a comparable for the second object.
16. The method of claim 14 , wherein the second object has a degree of separation from the first object that is less than a threshold degree of separation.
17. The method of claim 14 , further comprising determining a quality status of the network, wherein the quality status of the network is based on a structure of the network.
18. The method of claim 14 , wherein the first boundary comprises one of a neighborhood or a local area network.
19. The method of claim 14 , wherein the barrier comprises one of a firewall, a mountain, or a river.
20. The method of claim 14 , further comprising outputting information associated with the first, second, and third objects via one of an electronic mail or a network-based interface.