Weakly-Supervised Deep Learning for Domain Invariant Sentiment Classification

Abstract

The task of learning a sentiment classification model that adapts well to any target domain, different from the source domain, is a challenging problem. Majority of the existing approaches focus on learning a common representation by leveraging both source and target data during training. In this paper, we introduce a two-stage training procedure that leverages weakly supervised datasets for developing simple lift-and-shift-based predictive models without being exposed to the target domain during the training phase. Experimental results show that transfer with weak supervision from a source domain to various target domains provides performance very close to that obtained via supervised training on the target domain itself. CCS CONCEPTS• Computing methodologies → Supervised learning by classification; Semi-supervised learning settings; Neural networks.

Sleeping Lion or Sick Man? Machine Learning Approaches to Deciphering Heterogeneous Images of Chinese in North America

“…Given the binary nature of these labels, they only suggest whether a review is positive or negative, not how positive or negative it is. Scholars find that the BERT and related sentiment-analysis models trained on the IMDb generalize well to other datasets and contexts (Xie et al 2019, Kayal, Singh andGoyal 2020). When Kayal et al (2020) generalized the BERT model fine-tuned on the IMDb to sentiment analysis based on various Amazon, Yelp, weather, and scientific reviews, their corresponding F1 scores, defined as the harmonic mean of the precision and recall (sensitivity), varied from 80.5 percent to 92.5 percent (the average is 86.5 percent).…”

Understanding the Role of Affect Dimensions in Detecting Emotions from Tweets: A Multi-task Approach

“…Russel and Mehrabian's VAD model [23] for instance interprets emotions as points in a 3-D space with Valence (degree of pleasure or displeasure), Arousal (degree of calmness or excitement), and Dominance (degree of authority or submission) being the three orthogonal dimensions. Accordingly, the literature on text-based emotion analysis can be broadly divided into coarse-grained classification systems [10,[12][13][14]28] and fine-grained regression systems [22,24,29,30]. Although a coarse-grained approach is better-suited for the task of detecting emotions from tweets as observed in [4], prior works fail to exploit the direct correlation between the two models of emotion representation for finer interpretation.…”