+ All Categories
Home > Documents > Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Date post: 17-Dec-2015
Category:
Upload: helena-long
View: 216 times
Download: 0 times
Share this document with a friend
Popular Tags:
15
Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09
Transcript
Page 1: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Joint Sentiment/Topic Model for Sentiment Analysis

Chenghua Lin & Yulan HeCIKM09

Page 2: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Main Idea

This paper proposes a novel probabilistic modeling framework based on Latent Dirichlet Allocation (LDA), called joint sentiment/topic model (JST), which detects sentiment and topic simultaneously from text.

Page 3: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Related Works

• Sentiment classification based on Machine Learning (e.g. supervised) requires a large amount of human annotation

• Sentiment classification model trained in one domain cannot work well in another domain

• topic/feature detection and sentiment classification are often performed separately, which ignores their mutual dependence.– e.g. ‘unpredictable steering’:

• Negative in automobile review• Positive in movie review

Page 4: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

JST model

• 1. Fully unsupervised. No need for human annotation

• 2. Detect sentiment/topic simultaneously by considering their mutual relation

Page 5: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

LDA vs. JST

LDA• Two Matrices:

– D × T distribution: θ– T × W distribution: φ

JST• Three Matrices:

– D × S distribution: π– D × S × T distribution: θ– D × S × W distribution: φ

Page 6: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

LDA vs. JST (cont.)

LDA JST

Page 7: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Process of JST

Page 8: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Incorporating Model Priors

• One of the directions for improving the sentiment detection accuracy is to incorporate prior information or subjectivity lexicon (i.e., words bearing positive or negative polarity), which can be obtained in many different ways.– Paradigm word list– Mutual information– Full subjectivity lexicon– Filtered subjectivity lexicon

Page 9: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Experiment

• Sentiment Classification: Only consider two sentiment labels, i.e. positive or negative

• Topic Extraction

Page 10: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Sentiment Classification

Page 11: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Sentiment Classification (cont.)

Page 12: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Summary 1

• 1. Classification performance of JST is very close to the best performance of ML but save a lot of annotation work.

• 2. topic information indeed helps in sentiment classification as the JST model with the mixture of topics consistently outperforms a simple LDA model ignoring the mixture of topics.

Page 13: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Topic Extraction

Page 14: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Summary 2:

• Manually examining the data reveals that the terms that seem not conveying sentiments under these two topics in fact appear in the context expressing positive sentiments. The above analysis illustrates the effectiveness of JST in extracting mixture of topics from a corpus.

Page 15: Joint Sentiment/Topic Model for Sentiment Analysis Chenghua Lin & Yulan He CIKM09.

Conclusion

• 1. presented a joint sentiment/topic (JST) model which can detect document level sentiment and extract mixture of topics from text simultaneously.

• 2. fully unsupervised, thus provides more flexibilities and can be easier adapted to other domain.

• 3. yield competitive performance in document level sentiment classification compared other existing supervised approaches

• 4. discovered topics that corresponds to positive/negative sentiment


Recommended