| CPC G06F 16/9537 (2019.01) [G06F 16/9535 (2019.01)] | 20 Claims |
|
1. A computer-implemented method for selecting and serving local information items in two stages, comprising:
receiving a candidate information item;
in a first stage of the two stages, assessing, using a machine-trained localness-determining system, whether the candidate information item contains local content, the local content being content that matches a kind of information known to be selectively consumed within local geographic regions, and selectively ignored in regions outside of the respective local geographic regions, wherein the localness-determining system performs the assessing by: extracting feature information from the candidate information item using a set of feature-extracting processes; and mapping the feature information to a binary result that indicates whether the candidate item information contains local content without resolving a particular location and bounds associated with the candidate information item;
in a second stage of the two stages, upon determining that the information item contains local content, determining, using a scope-determining system, a particular geographic region associated with the information item, expressed by item-region information;
receiving an indication of a consumer location associated with a particular consumer;
determining a consumer region based on the consumer location of the particular consumer, the consumer region being expressed as consumer-region information; and
comparing the consumer-region information with the item-region information to produce a comparison result, and based on the comparison result, determining whether to serve the information item to the particular consumer.
|