Emotion Detection with Neural Personal Discrimination

Abstract

There have been a recent line of works to automatically predict the emotions of posts in social media. Existing approaches consider the posts individually and predict their emotions independently. Different from previous researches, we explore the dependence among relevant posts via the authors' backgrounds, since the authors with similar backgrounds, e.g., gender, location, tend to express similar emotions. However, such personal attributes are not easy to obtain in most social media websites, and it is hard to capture attributesaware words to connect similar people. Accordingly, we propose a Neural Personal Discrimination (NPD) approach to address above challenges by determining personal attributes from posts, and connecting relevant posts with similar attributes to jointly learn their emotions. In particular, we employ adversarial discriminators to determine the personal attributes, with attention mechanisms to aggregate attributes-aware words. In this way, social correlationship among different posts can be better addressed. Experimental results show the usefulness of personal attributes, and the effectiveness of our proposed NPD approach in capturing such personal attributes with significant gains over the state-of-the-art models.

UPB at SemEval-2021 Task 7: Adversarial Multi-Task Learning for Detecting and Rating Humor and Offense

“…Other works (Zhou et al, 2019;Spiliopoulou et al, 2020) used the multi-task technique alongside adversarial learning, but in other configurations than ASP-MTL. For example, Zhou et al (2019) employed a model that has a shared feature extractor, and then it is followed by branches for each task with an attention mechanism for the first layer from each branch. By adding new branches to the model, as well as adversarial learning, the F1-score increased when compared to the baseline model.…”