CATEGORIZING GEOGRAPHIC AREAS BASED ON USER INTERACTIONS

摘要

A system for categorizing geographic areas may include one or more processors and a memory. The one or more processors may receive a plurality of geographic locations each having been associated with one of a plurality of categories based at least in part on a plurality of user interactions. The one or more processors may associate a geographic area that encompasses the plurality of geographic locations with a category of the plurality of categories that has a highest commonality across the plurality of geographic locations. The one or more processors may receive an advertisement request that comprises an indication of one of the plurality of geographic locations that is encompassed by the geographic area, and provide an advertisement that is selected based at least in part on the category of the plurality of categories associated with the geographic area that encompasses the plurality of geographic locations.

主权项

1. A computer-implemented method for categorizing geographic areas, the method comprising:
determining, using one or more computing devices, a plurality of geographic locations each associated with one of a plurality of categories; determining, using the one or more computing devices, a plurality of geographic areas, each of the plurality of geographic areas encompassing at least two of the plurality of geographic locations, wherein the plurality of geographic areas are discrete; receiving, using the one or more computing devices, a plurality of user interactions with a map service; determining, using the one or more computing devices, a plurality of positive and negative outcomes based on the plurality of user interactions, wherein the positive outcomes are associated with successful searches and the negative outcomes are associated with unsuccessful searches, and wherein each of the plurality of outcomes is associated with one of the plurality of geographic locations; determining, using the one or more computing devices, a plurality of weights of the plurality of outcomes, wherein the plurality of weights indicate how strongly the plurality of outcomes are indicative of user intent; determining, using the one or more computing devices and with respect to each of the plurality of outcomes, one of a plurality of categories that relates to each of the plurality of outcomes; storing, using the one or more computing devices and with respect to each of the plurality of geographic areas, an association between a category of the plurality of categories that is associated with a largest number of the at least two of the plurality of geographic locations encompassed by each of the plurality of geographic areas, and each of the plurality of geographic areas, respectively; determining, using the one or more computing devices, clusters of proximal geographic locations that have a common category, wherein the determining clusters is based at least in part on the stored associations between each geographic area and its corresponding category; and providing, using the one or more computing devices, an advertisement that is selected based at least in part on the clusters of proximal geographic locations that are determined by the stored associations between the category of the plurality of categories and each of the plurality of geographic areas.