SYSTEM AND METHOD FOR SLANG SENTIMENT CLASSIFICATION FOR OPINION MINING

• US 20120259617A1
• Filed: 12/14/2011
• Published: 10/11/2012
• Est. Priority Date: 04/07/2011
• Status: Active Grant

• Current Assignee: Infosys Technologies Limited (Infosys Limited)
• Original Assignee: Infosys Technologies Limited (Infosys Limited)
• Inventors: Pisipati, Radha Krishna, Indukuri, Kishore Varma, Kantharaj, Manuel
• Granted Patent: US 10,474,752 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

The present disclosure describes a method of sentiment oriented slang for opinion mining. With increasing use of internet, many users can submit their review comments directly to the companies which can be automatically processed and summarized with critical issues from time to time and help the company get real time feedback from its customers. The method comprises, receiving at least one document comprising a plurality of sentiment oriented slang. The next step of the method comprises identifying the plurality of sentiment oriented slang in the at least one document. Further, a polarity score of each of a slang word identified is determined and sentiment information is displayed on an output device as an output.

Claims

1. A method comprising:

receiving at least one document comprising a plurality of sentiment oriented slang;

identifying the plurality of sentiment oriented slang in the at least one document;

determining a polarity score of each of a slang word identified using a processor; and

displaying a sentiment information on an output device.

2. The method of claim 1 further comprising classifying the plurality of sentiment oriented slang to a subjective context.

3. The method of claim 1 further comprising classifying the plurality of sentiment oriented slang to an objective context.

4. The method of claim 2 further comprising identifying the slang word in the subjective context.

5. The method of claim 4 further comprising determining the polarity score of each of the slang word.

6. The method of claim 5 further comprising computing an overall polarity score of the at least one document.

7. The method of claim 6 wherein the overall polarity score is the weighted summation of all the polarity score.

8. The method of claim 1 wherein the at least one document is at least one or more of:

a blog;

a website; and

a text document.

9. The method of claim 1 wherein the step of identifying the plurality of sentiment oriented slang in the at least one document further comprises, accessing at least one database to refer to the plurality of sentiment oriented slang.

10. The method of claim 1 wherein the step of receiving at least one document comprising a plurality of sentiment oriented slang further comprises receiving the at least one document from a review database.

11. The method of claim 1 wherein the step of determining a polarity score of each of a slang word identified further comprises accessing the at least one database to refer to a value.

12. The method of claim of claim 9 wherein the at least one database is a slang dictionary.

13. The method of claim 1 further comprises identifying an occurrence of the slang word in the at least one document.

14. The method of claim 13 further comprises counting the number of occurrence of the slang word and storing the number of occurrence of the slang word.

15. The method of claim 11 wherein the value is a number of occurrences of each of the slang word.

16. The method of claim 1 wherein the polarity score can be at least one or more of:

a strong negative score S1;

a weak negative score S2;

a neutral score S3;

a weak positive score S4; and

a strong positive score S5.

17. The method of claim 1 wherein the sentiment information is a report based on the overall polarity score of the at least one document.

18. The method of claim 1 wherein the output device can be at least one or more of a:

a computer monitor; and

a handheld device.

19. A system comprising:

a data gathering module configured to receive at least one input from at least one database;

a data preparation module configured to receive an input from the data gathering module, the data preparation module transforms the input received from the data gathering module;

a sentiment classification module configured to receive input from the data preparation module, the sentiment classification module further configured to identify a plurality of sentiment oriented slang in at least one document and calculate a polarity score of each slang word; and

an output module operable to communicate with the sentiment classification module, the output module displays an output on an output device.

20. The system of claim 19 wherein the sentiment classification module is configured to classify the plurality of sentiment oriented slang to a subjective context.

21. The system of claim 19 wherein the sentiment classification module is configured to classify the plurality of sentiment oriented slang to an objective context.

22. The system of claim 19 wherein the sentiment classification module is further configured to compute an overall polarity score, wherein the overall polarity score is the weighted summation of all polarity score.

23. The system of claim 19 wherein the output module is configured to display the output based on the overall polarity score of the at least one input.

24. The system of claim 19 wherein the polarity score can be at least one or more of:

a strong negative score S1;

a weak negative score S2;

a neutral score S3;

a weak positive score S4; and

a strong positive score S5.

25. The system of claim 19 wherein the at least one input can be at least one or more of:

a blog;

a website; and

a text document. 20

26. The system of claim 19 wherein the plurality of sentiment oriented slang is identified by referring to a slang dictionary.

27. The system of claim 19 wherein the sentiment classification module is further configured to count an occurrence the plurality of sentiment oriented slang.

28. The system of claim 22 wherein the polarity score is computed by referring to a value, wherein the value is the number of occurrences of each slang word.

29. The system of claim 19 wherein the at least one database is a review database.

30. The system of claim 19 wherein an output is a sentiment information based on the overall polarity score of the at least one input.

31. The system of claim 19 wherein the data preparation module transforms the input to a common format by normalizing the data.

32. The system of claim 19 wherein the output device can be at least one or more of a:

a computer monitor; and

a handheld device.

33. The system of claim 19 wherein the sentiment classification module is configured to recomputed the polarity score for modifications in the at least one input.

34. A computer program product, comprising a machine-accessible medium having instructions encoded thereon for enabling a processor to perform the operations of:

program code adapted for receiving at least one document comprising a plurality of sentiment oriented slang;

program code adapted for identifying the plurality of sentiment oriented slang in the at least one document;

program code adapted for determining a polarity score of each of a slang word identified; and

program code adapted for displaying a sentiment information on an output device.

35. The computer program product of claim 34, further comprising program code adapted for classifying the plurality of sentiment oriented slang to a subjective context.

36. The computer program product of claim 34, further comprising program code adapted for classifying the plurality of sentiment oriented slang to an objective context.

37. The computer program product of claim 34, further comprising program code adapted for comprising identifying the slang word in the subjective context.

38. The computer program product of claim 34, further comprising program code adapted for determining the polarity score of the slang word.