Identifying establishments in images

A computer system for identifying an establishment in a geo-tagged image, comprising:

• a non-transitory computer-readable storage medium comprising executable computer program code for:
• recognizing a text string in the geo-tagged image,
• identifying an establishment near a geographic location specified by a geo-tag of the geo-tagged image,
• extracting a textual phrase from information associated with the establishment near the geographic location,
• comparing the text string with the textual phrase to derive a match, and
• identifying the establishment near the geographic location as the establishment in the geo-tagged image based on the match;
• and
• a processor for executing the executable computer program code.

A non-transitory computer-readable storage medium storing executable computer program instructions for identifying an establishment in a geo-tagged image, the computer program instructions comprising instructions for:

• recognizing a text string in the geo-tagged image;
• identifying an establishment near a geographic location specified by a geo-tag of the geo-tagged image;
• extracting a textual phrase from information associated with the establishment;
• comparing the text string with the textual phrase to derive a match;
• and
• identifying the establishment near the geographic location as the establishment in the geo-tagged image based on the match.

A computer-implemented method of identifying an establishment in a geo-tagged image, comprising:

• recognizing at least one text string in the geo-tagged image;
• identifying at least one establishment near a geographic location specified by a geo-tag of the geo-tagged image;
• extracting at least one textual n-gram from information associated with the at least one establishment near the geographic location;
• comparing the at least one text string with the at least one textual n-gram to derive one or more matches;
• assigning a weight for a match based on one or more of the following factors:
• an information field from which an a textual n-gram associated with the match is extracted,
• a confidence measuring an accuracy of a text string associated with the match,
• a prevalence of the textual n-gram associated with the match in a language associated with the textual n-gram, and
• a prevalence of the textual n-gram associated with the match in a geographic area associated with the geo-tagged image;
• generating a score for an establishment near the geographic location based on one or more of the following factors:
• a number of matches associated with textual n-grams extracted from information associated with that establishment,
• a weight assigned for a match associated with an textual n-gram extracted from information associated with that establishment, and
• a distance between the geographic location specified by the geo-tag of the geo-tagged image and a location of that establishment;
• and
• identifying an establishment near the geographic location as the establishment in the geo-tagged image based on the score.