Stance Inference in Twitter through Graph Convolutional Collaborative Filtering Networks with Minimal Supervision
E. Zhiwei Zhou. Companion Proceedings of the ACM Web Conference 2023, page 1030--1038. L3S Research Center, (April 2023)
DOI: 10.1145/3543873.3587640
Abstract
Social Media (SM) has become a stage for people to share thoughts, emotions, opinions, and almost every other aspect of their daily lives. This abundance of human interaction makes SM particularly attractive for social sensing. Especially during polarizing events such as political elections or referendums, users post information
and encourage others to support their side, using symbols such as hashtags to represent their attitudes. However, many users choose not to attach hashtags to their messages, use a diferent language, or show their position only indirectly. Thus, automatically identifying their opinions becomes a more challenging task. To uncover
these implicit perspectives, we propose a collaborative fltering model based on Graph Convolutional Networks that exploits the textual content in messages and the rich connections between users and topics. Moreover, our approach only requires a small annotation efort compared to state-of-the-art solutions. Nevertheless, the proposed model achieves competitive performance in predicting individuals’ stances. We analyze users’ attitudes ahead of two constitutional referendums in Chile in 2020 and 2022. Using two large Twitter datasets, our model achieves improvements of 3.4% in recall
and 3.6% in accuracy over the baselines.
%0 Conference Paper
%1 zhou2023stance
%A Zhiwei Zhou, Erick Elejalde
%B Companion Proceedings of the ACM Web Conference 2023
%D 2023
%K myown from:zzhou
%P 1030--1038
%R 10.1145/3543873.3587640
%T Stance Inference in Twitter through Graph Convolutional Collaborative Filtering Networks with Minimal Supervision
%U https://doi.org/10.1145/3543873.3587640
%X Social Media (SM) has become a stage for people to share thoughts, emotions, opinions, and almost every other aspect of their daily lives. This abundance of human interaction makes SM particularly attractive for social sensing. Especially during polarizing events such as political elections or referendums, users post information
and encourage others to support their side, using symbols such as hashtags to represent their attitudes. However, many users choose not to attach hashtags to their messages, use a diferent language, or show their position only indirectly. Thus, automatically identifying their opinions becomes a more challenging task. To uncover
these implicit perspectives, we propose a collaborative fltering model based on Graph Convolutional Networks that exploits the textual content in messages and the rich connections between users and topics. Moreover, our approach only requires a small annotation efort compared to state-of-the-art solutions. Nevertheless, the proposed model achieves competitive performance in predicting individuals’ stances. We analyze users’ attitudes ahead of two constitutional referendums in Chile in 2020 and 2022. Using two large Twitter datasets, our model achieves improvements of 3.4% in recall
and 3.6% in accuracy over the baselines.
@inproceedings{zhou2023stance,
abstract = {Social Media (SM) has become a stage for people to share thoughts, emotions, opinions, and almost every other aspect of their daily lives. This abundance of human interaction makes SM particularly attractive for social sensing. Especially during polarizing events such as political elections or referendums, users post information
and encourage others to support their side, using symbols such as hashtags to represent their attitudes. However, many users choose not to attach hashtags to their messages, use a diferent language, or show their position only indirectly. Thus, automatically identifying their opinions becomes a more challenging task. To uncover
these implicit perspectives, we propose a collaborative fltering model based on Graph Convolutional Networks that exploits the textual content in messages and the rich connections between users and topics. Moreover, our approach only requires a small annotation efort compared to state-of-the-art solutions. Nevertheless, the proposed model achieves competitive performance in predicting individuals’ stances. We analyze users’ attitudes ahead of two constitutional referendums in Chile in 2020 and 2022. Using two large Twitter datasets, our model achieves improvements of 3.4% in recall
and 3.6% in accuracy over the baselines.},
added-at = {2024-02-16T11:40:50.000+0100},
author = {Zhiwei Zhou, Erick Elejalde},
biburl = {https://www.bibsonomy.org/bibtex/2ac26babce2cbcf15ed14aef122a25ca2/l3s},
booktitle = {Companion Proceedings of the ACM Web Conference 2023},
dnbtitleid = {571867820},
doi = {10.1145/3543873.3587640},
eventdate = {April 30 - May 04},
eventtitle = {Companion Proceedings of the ACM Web Conference 2023},
interhash = {800becd1904334b7850e22acdceb4128},
intrahash = {ac26babce2cbcf15ed14aef122a25ca2},
keywords = {myown from:zzhou},
language = {English},
month = {April},
organization = {L3S Research Center},
pages = {1030--1038},
school = {Leibniz University Hannover},
timestamp = {2024-02-16T11:40:50.000+0100},
title = {Stance Inference in Twitter through Graph Convolutional Collaborative Filtering Networks with Minimal Supervision},
url = {https://doi.org/10.1145/3543873.3587640},
year = 2023
}