A Token Level Multi-target Stance Detection Dataset

More and more people would like to express their opinions on the social media platform, such as Twitter, Sina Weibo, or Facebook. High-performance stance detection algorithm has become a meaningful and important application. The accuracy improvement of the stance detection model significantly relies on the quality of the training dataset. Previously, the traditional social media textual datasets with manual annotation are tagged on the sentence level mostly. It leads to lacking fine-grained analysis and generalization ability of the stance detection algorithm. Therefore, we propose a token level stance detection dataset with 2025 labeled tweets. The multiple targets of stances in the tweet are tagged at a token level. It uses the token level target annotations instead of using a list of hashtags to represent the stance target. The experiment shows that our dataset can improve the classification accuracy of the stance detection algorithm.

PDF
No code implementations yet. Submit your code now

Datasets


Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here