Point of interest tagging from social feeds

A computing device comprising:

• a computing processor;
• and a computer storage medium comprising computer-executable instructions stored thereon which, when executed by the computing processor cause the computing processor to perform electronic operations comprising:
• obtaining location-related social media messages posted from a plurality of users from a neighborhood of an entity, wherein the neighborhood comprises an area around the entity with a predetermined range;
• determining, from the obtained messages, one or more terms that are unique to the entity after filtering the obtained messages with one or more attributes that are unique to the entity in a local index of a search engine;
• individually determining one or more co-occurring terms within the one or more unique terms, wherein the one or more co-occurring terms are identified as being co-occurring based on a frequency of use in a social media network, wherein the frequency of use is based on use of the terms in the location-related social media messages having associated geo-location information that indicates posting within the neighborhood in addition to use of the terms in social media messages having an absence of associated geo-location information;
• enriching an entry for the entity in the local index using the one or more co-occurring terms as one or more search tags to label the entity in the local index, wherein the one or more search tags are used to label the entity and one or more other entities in the local index of the search engine;
• receiving a location-related search query, the location-related search query including a first and a second portion of text, wherein the first portion of text provides one or more search terms associated with the one or more search tags and wherein the second portion of text provides one or more location terms associated with the area around the entity;
• performing a search of the local index using the search engine, the search of the local index being performed using the one or more search tags and the one or more location terms, wherein the search identifies the area around the entity based on the one or more location terms, and wherein the search identifies the one or more co-occurring terms relevant to the entity based on the one or more search tags;
• retrieving search results associated with the entity from the search of the local index;
• and outputting the search results associated with the entity for display on a user interface.

One or more non-transitory computer storage media comprising computer-executable instructions stored thereon which, when executed by a processor, cause the processor to perform a method comprising:

• obtaining location-related social media messages posted from a plurality of users from within a neighborhood of an entity eligible for inclusion as an entry in a local index of a search engine, wherein the neighborhood comprises an area around the entity with a predetermined range;
• determining, from the obtained messages, one or more terms that are unique to the entity after filtering the obtained messages with one or more attributes that are unique to the entity in the local index of the search engine;
• individually determining one or more co-occurring terms within the one or more unique terms, wherein the one or more co-occurring terms are identified as being co-occurring based on a frequency of use in a social media network, wherein the frequency of use is based on use of the terms in the location-related social media messages having associated geo-location information that indicates posting within the neighborhood in addition to use of the terms in social media messages having an absence of associated geo-location information;
• enriching the entry for the entity using the one or more co-occurring terms as one or more search tags to label the entity in the local index of the search engine, wherein the one or more search tags are used to label the entity and one or more other entities in the local index of the search engine;
• receiving a location-related search query from a user device, the location-related search query including a first and a second portion of text, wherein the first portion of text provides one or more search terms associated with the one or more search tags and wherein the second portion of text provides one or more location terms associated with the area around the entity;
• extracting search terms from the location-related search query;
• performing a search of the local index using the search engine, the search of the local index being performed using the one or more search tags of the entity and the one or more location terms, and comparing the extracted search terms with search tags of respective entities in the local index, wherein the search identifies the area around the entity based on the one or more location terms, and wherein the search identifies the one or more co-occurring terms relevant to the entity based on the one or more search tags;
• retrieving search results from the search of the local index, the search results comprising a matching entity together with the location-related social media messages associated with the matching entity;
• and
• providing the search results to the user device for displaying the location-related social media messages in a ranking based at least in part on one or more of whether respective authors of the location-related social media messages have verified accounts, or respective numbers of points of interest (POI) metadata tags of the matching entity that are found in the location-related social media messages;
• and
• outputting the search results for display on a user interface of the user device according to the ranking.