AI helps you reading Science
AI generates interpretation videos
AI extracts and analyses the key points of the paper to generate videos automatically
AI parses the academic lineage of this thesis
AI extracts a summary of this paper
We extract sentiment relations between tweets based on social theories, and model the relations using graph Laplacian, which is employed as a regularization to a sparse formulation
Exploiting social relations for sentiment analysis in microblogging
WSDM, pp.537-546, (2013)
Microblogging, like Twitter and Sina Weibo, has become a popular platform of human expressions, through which users can easily produce content on breaking news, public events, or products. The massive amount of microblogging data is a useful and timely source that carries mass sentiment and opinions on various topics. Existing sentiment a...More
PPT (Upload PPT)
- Microblogging services are extensively used to share information or opinions in various domains.
- The sheer volumes of microblogging data present opportunities and challenges for sentiment analysis of these noisy and short texts.
- Sentiment analysis has been extensively studied for product and movie reviews , which differ substantially from microblogging data.
- When composing a microblogging message, users may use or coin new abbreviations or acronyms that seldom appear in conventional text documents.
- Messages like “It is coooooool”, “OMG :-(”, are intuitive and popular in microblogging, but some are not formal words.
- Existing methods [2, 8] rely on pre-defined sentiment vocabularies , which are highly domain-specific
- Microblogging services are extensively used to share information or opinions in various domains
- We introduce our formulation to utilize social relations for sentiment analysis, and answer the question: “How can social relations be explicitly integrated into the sentiment classification framework?”
- We propose a novel sociological approach (SANT ) to handle networked texts in microblogging
- We extract sentiment relations between tweets based on social theories, and model the relations using graph Laplacian, which is employed as a regularization to a sparse formulation
- Experimental results show that the user-centric social relations are helpful for sentiment classification of microblogging messages
- Empirical evaluations demonstrate that our framework significantly outperforms the representative sentiment classification methods on two real-world datasets, and SANT achieves consistent performance for different sizes of training data, a useful feature for sentiment classification
- We present empirical evaluation results to assess the effectiveness of our proposed framework, and answer the question: “Can social relations improve sentiment classification of microblogging messages?”.
- The authors first present the finding of comparing SANT with the classical text-based sentiment classification methods, and with the models that incorporate social relations in sentiment classification.
- In the first set of experiments, the authors use classification accuracy as the performance metric, and compare the proposed framework SANT with following text-based methods: LS Lasso MinCuts LexRatio SANT D10% D25% D50% D100% 0.670 (N.A.) 0.704 (N.A.)
- Empirical evaluations demonstrate that the framework significantly outperforms the representative sentiment classification methods on two real-world datasets, and SANT achieves consistent performance for different sizes of training data, a useful feature for sentiment classification.
- CONCLUSIONS AND FUTURE WORK
Different from texts in traditional media, microblogging texts are noisy, short, and embedded with social relations, which presents challenges to sentiment analysis.
- The proposed method can utilize sentiment relations between messages to facilitate sentiment classification and effectively handle noisy Twitter data.
- Experimental results show that the user-centric social relations are helpful for sentiment classification of microblogging messages.
- Empirical evaluations demonstrate that the framework significantly outperforms the representative sentiment classification methods on two real-world datasets, and SANT achieves consistent performance for different sizes of training data, a useful feature for sentiment classification
- Table1: Statistics of the Datasets
- Table2: Sentiment Classification Accuracy on STS Dataset
- Table3: Sentiment Classification Accuracy on OMD Dataset
- Recently, sentiment analysis on microblogging, which is considered to be an opinion-rich resource, has gained huge popularity and attracted researchers from many disciplines [3, 18, 20, 30]. Bollen et al  proposed to measure the sentiments on Twitter over time, and compared the correlation between sentiments and major events, including the stock market, crude oil prices, elections and Thanksgiving. Also, Kim et al  examined a tweet dataset about Michael Jackson’s death to gain insight into how emotion is expressed on Twitter. O’Connor et al  used sentiment analysis to automatically label the sentiments of tweets about politicians, and found strong correlation between the aggregated sentiment and the manually collected poll ratings.
Sentiment classification has been studied for years on various text corpus, like newspaper articles , movie reviews , and product reviews [5, 11, 23]. The basic idea of the methods is to build a sophisticated feature space, which can effectively represent the sentiment status of the texts. Existing methods, which are designed for traditional i.i.d. text data, cannot effectively make use of the abundant social relation information contained in microblogging. Following the methods for traditional texts, there are some existing efforts in the community on the microblogging data. Alec et al  presented the results of machine learning algorithms for classifying the sentiments of Twitter messages using distant supervision. Barbosa and Feng  explored the linguistic characteristics of how tweets are written and the metainformation of words for sentiment classification. The ideas of the methods are consistent with traditional ones, ignoring the social relation information.
- This work is, in part, supported by ONR (N000141110527) and (N000141010091)
- R. Abelson. Whatever became of consistency theory? Personality and Social Psychology Bulletin, 1983.
- L. Barbosa and J. Feng. Robust sentiment detection on twitter from biased and noisy data. In Proceedings of ACL, 2010.
- J. Bollen, A. Pepe, and H. Mao. Modeling public mood and emotion: Twitter sentiment and socio-economic phenomena. CoRR, abs/0911.1583, 2009.
- F. Chung. Spectral graph theory. Number 92. Amer Mathematical Society, 1997.
- X. Ding and B. Liu. The utility of linguistic rules in opinion mining. In Proceedings of SIGIR, 2007.
- J. Fowler and N. Christakis. The dynamic spread of happiness in a large social network. BMJ: British medical journal, 2008.
- J. Friedman, T. Hastie, and R. Tibshirani. The elements of statistical learning, 2008.
- A. Go, R. Bhayani, and L. Huang. Twitter sentiment classification using distant supervision. Technical Report, Stanford, 2009.
- Q. Gu and J. Han. Towards feature selection in network. In Proceedings of CIKM, 2011.
- E. Hatfield, J. Cacioppo, and R. Rapson. Emotional contagion. Current Directions in Psychological Science, 1993.
- M. Hu and B. Liu. Mining and summarizing customer reviews. In Proceedings of SIGKDD, 2004.
- X. Hu and H. Liu. Text analytics in social media. Mining Text Data, pages 385–414, 2012.
- X. Hu, N. Sun, C. Zhang, and T.-S. Chua. Exploiting internal and external semantics for the clustering of short texts using world knowledge. In Proceedings of CIKM, pages 919–928, 2009.
- X. Hu, L. Tang, and H. Liu. Enhancing accessibility of microblogging messages using semantic knowledge. In Proceedings of the 20th CIKM, 2011.
- Y. Hu, A. John, D. Seligmann, and F. Wang. What were the tweets about? topical associations between public events and twitter feeds. ICWSM, 2012.
- Y. Hu, A. John, F. Wang, and S. Kambhampati. Et-lda: Joint topic modeling for aligning events and their twitter feedback. In AAAI, 2012.
- S. Ji and J. Ye. An accelerated gradient method for trace norm minimization. In ICML, 2009.
- S. Kamvar and J. Harris. We feel fine and searching the emotional web. In Proceedings of WSDM, pages 117–126, 2011.
- D. Keltner and G. Bonanno. A study of laughter and dissociation: Distinct correlates of laughter and smiling during bereavement. Journal of Personality and Social Psychology, page 687, 1997.
- E. Kim, S. Gilbert, M. Edwards, and E. Graeff. Detecting sadness in 140 characters: Sentiment analysis of mourning michael jackson on twitter. 2009.
- H. Kwak, C. Lee, H. Park, and S. Moon. What is twitter, a social network or a news media? In Proceedings of WWW, 2010.
- K. Lewis, M. Gonzalez, and J. Kaufman. Social selection and peer influence in an online social network. PNAS, pages 68–72, 2012.
- B. Liu. Sentiment analysis and opinion mining. Synthesis Lectures on Human Language Technologies, pages 1–167, 2012.
- J. Liu, S. Ji, and J. Ye. Multi-task feature learning via efficient l 2, 1-norm minimization. In Proceedings of UAI, 2009.
- P. Marsden and N. Friedkin. Network studies of social influence. Sociological Methods & Research, 1993.
- M. McPherson, L. Smith-Lovin, and J. Cook. Birds of a feather: Homophily in social networks. Annual review of sociology, pages 415–444, 2001.
- M. Miller, C. Sathi, D. Wiesenthal, J. Leskovec, and C. Potts. Sentiment flow through hyperlink networks. In Proceedings of ICWSM, 2011.
- A. Mislove, S. Lehmann, Y. Ahn, J. Onnela, and J. Rosenquist. Pulse of the nation: Us mood throughout the day inferred from twitter. 2010.
- Y. Nesterov and I. Nesterov. Introductory lectures on convex optimization: A basic course. 2004.
- B. O Connor, R. Balasubramanyan, B. Routledge, and N. Smith. From tweets to polls: Linking text sentiment to public opinion time series. In Proceedings of ICWSM, 2010.
- B. Pang and L. Lee. A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts. In Proceedings of ACL, 2004.
- B. Pang and L. Lee. Opinion mining and sentiment analysis. Foundations and Trends in Information Retrieval, pages 1–135, 2008.
- B. Pang, L. Lee, and S. Vaithyanathan. Thumbs up?: sentiment classification using machine learning techniques. In Proceedings of EMNLP, 2002.
- D. Shamma, L. Kennedy, and E. Churchill. Tweet the debates: understanding community annotation of uncollected sources. In Proceedings of the first SIGMM workshop on Social media, pages 3–10. ACM, 2009.
- M. Speriosu, N. Sudan, S. Upadhyay, and J. Baldridge. Twitter polarity classification with label propagation over lexical links and the follower graph. In Proceedings of the First Workshop on Unsupervised Learning in NLP, 2011.
- C. Tan, L. Lee, J. Tang, L. Jiang, M. Zhou, and P. Li. User-level sentiment analysis incorporating social networks. In Proceedings of SIGKDD, 2011.
- J. Tang and H. Liu. Feature selection with linked data in social media. In SDM, 2012.
- J. Tang, X. Wang, H. Gao, X. Hu, and H. Liu. Enriching short texts representation in microblog for clustering. Frontiers of Computer Science.
- J. Wiebe, T. Wilson, and C. Cardie. Annotating expressions of opinions and emotions in language. Language Resources and Evaluation, 39:165–210, 2005.
- T. Wilson, J. Wiebe, and P. Hoffmann. Recognizing contextual polarity in phrase-level sentiment analysis. In Proceedings of HLT and EMNLP, 2005.
- X. Zhao, G. Li, M. Wang, J. Yuan, Z.-J. Zha, Z. Li, and T.-S. Chua. Integrating rich information for video recommendation with multi-task rank aggregation. In Proceedings of ACM MM, 2011.