User-level sentiment analysis incorporating social networks
Chenhao Tan, Lillian Lee, Jie Tang, Long Jiang, Ming Zhou, and Ping Li
Proceedings of KDD, pp. 1397--1405, 2011. Poster paper. (The notification letter gave the aggregate acceptance rate for oral presentation plus posters as 17.5%)

We show that information about social relationships can be used to improve user-level sentiment analysis. The main motivation behind our approach is that users that are somehow "connected" may be more likely to hold similar opinions; therefore, relationship information can complement what we can extract about a user's viewpoints from their utterances. Employing Twitter as a source for our experimental data, and working within a semi-supervised framework, we propose models that are induced either from the Twitter follower/followee network or from the network in Twitter formed by users referring to each other using "@" mentions. Our transductive learning results reveal that incorporating social-network information can indeed lead to statistically significant sentiment classification improvements over the performance of an approach based on Support Vector Machines having access only to textual features

@inproceedings{Tan+al:11a, author = {Chenhao Tan and Lillian Lee and Jie Tang and Long Jiang and Ming Zhou and Ping Li}, title = {User-level sentiment analysis incorporating social networks}, year = {2011}, pages = {1397--1405}, booktitle = {Proceedings of KDD} }


This material is based upon work supported in part by Chinese National Key Foundation Research 60933013 and 61035004, a Google Research Grant, a grant from Microsoft, ONR YIP-N000140910911, China's National Hightech R&D Program 2009AA01Z138, Natural Science Foundation of China 61073073, US NSF DMS-0808864 and IIS-0910664, and a Yahoo! Faculty Research and Engagement Award. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or other sponsors.