Exploiting Social Network Structure for Person-to-Person Sentiment Analysis

By Robert West, Hristo S. Paskov, Jure Leskovec, Christopher Potts. 2014

233 citations recorded by the source corpus.

Person-to-person evaluations are prevalent in all kinds of discourse and important for establishing reputations, building social bonds, and shaping public opinion. Such evaluations can be analyzed separately using signed social networks and textual sentiment analysis, but this misses the rich interactions between language and social context. To capture such interactions, we develop a model that predicts individual A’s opinion of individual B by synthesizing information from the signed social network in which A and B are embedded with sentiment analysis of the evaluative texts relating A to B. We prove that this problem is NP-hard but can be relaxed to an efficiently solvable hinge-loss Markov random field, and we show that this implementation outperforms text-only and network-only versions in two very different datasets involving community-level decision-making: the Wikipedia Requests for Adminship corpus and the Convote U.S. Congressional speech corpus.

Subfields: Artificial Intelligence, Sociology and Political Science

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xing Xie
2. Hongwei Wang
3. Fuzheng Zhang
4. Minyi Guo
5. Jia Deng
6. Li Fei-Fei
7. Jonathan Krause
8. Lise Getoor

In Sociology and Political Science

1. Xing Xie
2. Jia Deng
3. Christos Faloutsos
4. Lise Getoor
5. Timnit Gebru
6. Stephen H. Bach
7. Srijan Kumar
8. Bert Huang

Citations in the corpus

• SHINE: Signed Heterogeneous Information Network Embedding for Sentiment Link Prediction (2017, background): “• Wiki-RfA: Wikipedia Requests for Adminship [30] is a signed network with 10,835 nodes and 159,388 edges, corresponding to votes cast by Wikipedia uses in election for promoting individuals to the role of administrator.”
• WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, neutral): “…(Koenecke et al., 2020; Tatman, 2017) and personalized recommender systems (Patro et al., 2020), tools used for analysis of individuals such as sentiment classification in computational social science (West et al., 2014) and user analytics (Lau et al., 2014), and applications in federated learning.”
• Fine-Grained Car Detection for Visual Census Estimation (2017, background): “However, the emergence of large, diverse sets of data generated by people has enabled computer scientists and computational sociologists to gain interesting insights by analyzing massive user texts and social networks (West et al. 2014; Cheng, Danescu-Niculescu-Mizil, and Leskovec 2014).”
• Edge Weight Prediction in Weighted Signed Networks (2016, background): “Wikipedia Requests for Adminship (RfA) network [9] is a signed network among Wikipedia users where each edge (u, v) has a weight corresponding to the vote of user u (-1 for negative, 0 for neutral, and 1 for positive) towards user v to become an administrator.”
• Hinge-Loss Markov Random Fields and Probabilistic Soft Logic (2015, background): “, 2014) and trust (Huang et al., 2013; West et al., 2014) prediction in social networks.”
• Hinge-Loss Markov Random Fields and Probabilistic Soft Logic (2015, background): “…Deng and Wiebe, 2015; Ebrahimi et al., 2016), automobile-traffic modeling (Chen et al., 2014), recommender systems (Kouki et al., 2015), information retrieval (Alshukaili et al., 2016), and predicting attributes (Li et al., 2014) and trust (Huang et al., 2013; West et al., 2014) in social networks.”