Performing sentiment analysis

Patent metadata

Field	Value
Publication number	US-2016307114-A1
Application number	US-201615044351-A
Country	US
Kind code	A1
Filing date	Feb 16, 2016
Priority date	Oct 24, 2011
Publication date	Oct 20, 2016
Grant date	—

Abstract

There is provided a computer-implemented method of performing sentiment analysis. An exemplary method comprises performing a first sentiment analysis on microblogging data based on a method using an opinion lexicon. The method also includes training a classifier using training data from the first sentiment analysis. Additionally, the method includes identifying a new opinion term in the microblogging data by performing a statistical test. The new opinion terms are not in the opinion lexicon. The method also includes identifying new microblogging data based on the new opinion term. Further, the method includes performing a second sentiment analysis on the new microblogging data using the classifier.

First claim

1 . A method comprising:
performing a first sentiment analysis on microblogging data using an opinion lexicon; adding a new opinion term to the opinion lexicon, the new opinion term identified in the microblogging data using results of the first sentiment analysis; identifying additional microblogging data, based on the opinion lexicon to which the new opinion term has been added; performing a second sentiment analysis on the additional microblogging data to generate training data; and training a classifier using the training data.

2 . The method of claim 1 , further comprising identifying the new opinion term using the results of the first sentiment analysis.

3 . The method of claim 2 , wherein identifying the new opinion term using the results of the first sentiment analysis comprises performing a statistical test on the results of the first sentiment analysis.

4 . The method of claim 1 , wherein the training data is second training data, wherein performing the first sentiment analysis generates first training data, and wherein the method further comprises:
training the classifier using the first training data, wherein training the classifier using the second training data comprises retraining the classifier as has been trained using the first training data, using the second training data.

5 . The method of claim 1 , wherein the opinion lexicon that is used to perform the first sentiment analysis includes non-domain-specific opinion terms, and wherein the new opinion term is domain-specific to the microblogging data.

6 . A non-transitory machine-readable storage medium encoded with instructions executable by at least one processor, the machine-readable storage medium comprising instructions to:
perform a first sentiment analysis on data using a lexicon; identify a new term in the data using results of the first sentiment analysis; add the new term to the lexicon; identify additional data, based on the lexicon to which the new term has been added; perform a second sentiment analysis on the additional data to generate training data; and train a classifier using the training data.

7 . The non-transitory machine-readable storage medium of claim 6 , wherein the instructions to identify the new term using the results of the first sentiment analysis comprise instructions to perform a statistical test on the results of the first sentiment analysis.

8 . The non-transitory machine-readable storage medium of claim 7 , wherein the statistic test comprises a Pearsons chi-square method, and wherein the new opinion term has a Pearsons chi-square value greater than a threshold.

9 . The non-transitory machine-readable storage medium of claim 6 , wherein the training data is second training data, wherein the instructions to perform the first sentiment analysis generate first training data, and wherein the instructions further comprises instructions to:
train the classifier using the first training data, wherein the instructions to train the classifier using the second training data retrain the classifier as has been trained using the first training data, using the second training data.

10 . The non-transitory machine-readable storage medium of claim 6 , wherein the lexicon that is used to perform the first sentiment analysis includes non-domain-specific terms,
and wherein the new term is domain-specific to the data.

11 . A system comprising:
a processor; a memory device storing instructions executable by the processor to:
identify new microblogging data, based on an opinion lexicon that has been improved by a new opinion term identified in existing microblogging data;
perform a sentiment analysis on the new microblogging data to generate training data; and
train a classifier using the training data.

12 . The system of claim 11 , wherein the sentiment analysis is a second sentiment analysis, and wherein the processor is to identify the new microblogging data by performing a first sentiment analysis on the existing microblogging data using the opinion lexicon prior to improvement.

13 . The system of claim 12 , wherein the processor is to identify the new microblogging data by further performing a statistical test on results of the first sentiment analysis.

14 . The system of claim 12 , wherein the training data is second training data, performing the first sentiment analysis generates first training data, and wherein the processor is further to:
train the classifier using the first training data, wherein the processor is to train the classifier using the second training data by retraining the classifier as has been trained using the first training data, using the second training data.

15 . The system of claim 11 , wherein the lexicon prior to improvement includes non-domain-specific opinion terms,
and wherein the new opinion term is domain-specific to the existing microblogging data.

Assignees

• Hewlett Packard Entpr Dev Lp

Inventors

• Ghosh Riddhiman
• Dekhil Mohamed E
• Zhang Lei

Classifications

• G06Q30/02

  Marketing; Price estimation or determination; Fundraising · CPC title

• G06F16/355

  Creation or modification of classes or clusters · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G06F17/2785

  Physics · mapped topic

• G06N99/005 Primary

  Physics · mapped topic