METHOD AND SYSTEM FOR EXTRACTING AND CLASSIFYING GEOLOCATION INFORMATION UTILIZING ELECTRONIC SOCIAL MEDIA

• US 20130086072A1
• Filed: 10/03/2011
• Published: 04/04/2013
• Est. Priority Date: 10/03/2011
• Status: Abandoned Application

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Peng, Wei, Sun, Tong, DeRoller, Matthew, Jaiswal, Anuj
• Application Number:
  US13/251,731
• Publication Number:
  US 20130086072A1
• Time in Patent Office: Days
• US Class Current: 707/743
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Abstract

Methods, systems and processor-readable media for extracting and classifying location information utilizing social media messages and/or data thereof. The social media messages can be sampled from a social media database and the messages filtered based on a heuristic rule. A geolocation entity from the unstructured social media messages can be extracted utilizing a geolocation entity extracting module. The messages with the geoentities can be uploaded onto a crowd sourcing platform to manually annotate the messages with a label. A text classification model can be built and learned from the label utilizing a machine learning algorithm and the messages can be classified by a location classifier in order to extract the user location. The user location can then be transformed into a geocode so that a spatial search can be enabled and the distance between the locations can be easily calculated.

Claims

1. A method for extracting and classifying user geolocation information, said method comprising:

sampling a plurality of social media messages comprising text, from a social media database in order to thereafter filter said plurality of social media messages based on a heuristic rule utilizing a heuristic message filtering module and generate at least one social media message filtered from said plurality of social media messages via said heuristic message filtering module;

extracting a geolocation entity from said at least one social media message utilizing a geolocation entity-extracting module;

uploading said at least one message onto a crowd sourcing platform to manually annotate said at least one social media message with a label; and

training a text classification model from said label utilizing a machine-learning algorithm in order to thereafter classify said at least one social medial message by a location classifier and extract location data.

2. The method of claim 1 further comprising transforming said location data into a geocode in order to enable a spatial search and calculate a distance between said locations.

3. The method of claim 1 further comprising filtering said plurality of social media messages in order to obtain a plurality of location messages and to reduce noisy data.

4. The method of claim 1 further comprising performing said geolocation entity extraction utilizing at least one of the following types of rules:

a geographic dictionary.

5. The method of claim 1 further comprising analyzing said plurality of user location messages in order to classify said plurality of user location messages into a past location, a current location, and a future location.

6. The method of claim 1 wherein said machine learning algorithm comprises at least one of the following types of algorithms:

a maximum entropy;

Naive Bayes; and

a support vector machine.

7. The method of claim 1 further comprising generating a text feature for said location classification by masking said location and including a bi-gram.

8. The method of claim 1 further comprising generating a text feature for said location classification by not removing a stop word and including a feature selection utilizing an information gain.

9. A system for extracting and classifying user geolocation information, said system comprising:

a processor;

a data bus coupled to said processor; and

a computer-usable medium embodying computer code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for;

sampling a plurality of social media messages comprising text, from a social media database in order to thereafter filter said plurality of social media messages based on a heuristic rule utilizing a heuristic message filtering module and generate at least one social media message filtered from said plurality of social media messages via said heuristic message filtering module;

extracting a geolocation entity from said at least one social media message utilizing a geolocation entity-extracting module;

uploading said at least one message onto a crowd sourcing platform to manually annotate said at least one social media message with a label; and

training a text classification model from said label utilizing a machine-learning algorithm in order to thereafter classify said at least one social medial message by a location classifier and extract location data.

10. The system of claim 9 wherein said instructions are further configured for transforming said location data into a geocode in order to enable a spatial search and calculate a distance between said locations.

11. The system of claim 9 wherein said instructions are further configured for filtering said plurality of social media messages in order to obtain a plurality of location messages and to reduce noisy data.

12. The system of claim 9 wherein said instructions are further configured for performing said geolocation entity extraction utilizing at least one of the following types of rules:

a geographic dictionary.

13. The system of claim 9 wherein said instructions are further configured for analyzing said plurality of user location messages in order to classify said plurality of user location messages into a past location, a current location, and a future location.

14. The system of claim 9 wherein said machine learning algorithm comprises at least one of the following types of algorithms:

a maximum entropy;

Naive Bayes; and

a support vector machine.

15. The system of claim 9 wherein said instructions are further configured for generating a text feature for said location classification by masking said location and including a bi-gram.

16. The system of claim 9 wherein said instructions are further configured for generating a text feature for said location classification by not removing a stop word and including a feature selection utilizing an information gain.

17. A processor-readable medium storing code representing instructions to cause a processor to perform a process to extract and classify user geolocation information, said code comprising code to:

sample a plurality of social media messages comprising text, from a social media database in order to thereafter filter said plurality of social media messages based on a heuristic rule utilizing a heuristic message filtering module and generate at least one social media message filtered from said plurality of social media messages via said heuristic message filtering module;

extract a geolocation entity from said at least one social media message utilizing a geolocation entity-extracting module;

upload said at least one message onto a crowd sourcing platform to manually annotate said at least one social media message with a label; and

train a text classification model from said label utilizing a machine-learning algorithm in order to thereafter classify said at least one social medial message by a location classifier and extract location data.

18. The processor-readable medium of claim 17 further comprises code to transform said location data into a geocode in order to enable a spatial search and calculate a distance between said locations.

19. The processor-readable medium of claim 17 further comprises code to filter said plurality of social media messages in order to obtain a plurality of location messages and to reduce noisy data.

20. The processor-readable medium of claim 17 further comprises code to perform said geolocation entity extraction utilizing at least one of the following types of rules:

a geographic dictionary.