Exploiting Social Network Structure for Person-to-Person Sentiment Analysis
This addresses the challenge of integrating social context with language for more accurate sentiment analysis in domains like online communities and political discourse, representing a hybrid incremental advance.
The paper tackles the problem of predicting person-to-person opinions by developing a model that combines signed social network structure with textual sentiment analysis, proving the problem is NP-hard but providing an efficient relaxation. The implementation outperforms text-only and network-only baselines on Wikipedia Requests for Adminship and Convote U.S. Congressional speech datasets.
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.