Systems and methods for identifying geographic locations of social media content collected over social networks

A method, comprising:

• retrieving social media items from a social network in real time, wherein the retrieved social media items do not have IP address identifiers to identify their geographic locations;
• determining a geographic location for each of the social media items retrieved in real time by analyzing n-grams of the content of that social media item which is authored or originated from an author for information in this content that indicates the geographic location, wherein the social media item lacks latitude, longitude location information and the n-grams are a contiguous sequence of words of length N in the content of that social media item;
• determining a geographic location indication of the content using a feature, location probabilistic classifier that uses a training set of location-known content to predict a location for the content without geographic coordinates;
• tagging each of the social media items with the corresponding location information;
• in response to the determination of the geographic location, confirming the identified location from the n-grams analysis for each of the social media items by comparing the identified location of this social media item to locations of prior social media items posted or authored by the same author;
• and
• presenting a view of at least some of the geo-tagged social media items with their associated geographic locations to a user.

A non-transitory machine-readable medium containing executable program instructions which when executed by a data processing device cause the device to perform a method, the method comprising:

• retrieving social media items from a social network in real time, wherein the retrieved social media items do not have IP address identifiers to identify their geographic locations;
• determining a geographic location for each of the social media items retrieved in real time by analyzing n-grams of content of that social media item which is authored or originated from an author for information in this content that indicates the geographic location, wherein the social media item lacks latitude and longitude location information, the n-grams are a contiguous sequence of words of length N in the content of that social media item;
• determining a geographic location indication of the content using a feature, location probabilistic classifier that uses a training set of location-known content to predict a location for the content without geographic coordinates;
• tagging each of the social media items with the corresponding location information;
• in response to the determination of the geographic location, confirming the identified location from the n-grams analysis for each of the social media items by comparing the identified location of this social media item to locations of prior social media items posted or authored by the same author;
• and
• presenting a view of at least some of the geo-tagged social media items with their associated geographic locations to a user.