KinGDOM: Knowledge-Guided DOMain Adaptation for Sentiment Analysis
By Deepanway Ghosal, Devamanyu Hazarika, Navonil Majumder, Abhinaba Roy, Soujanya Poria, Rada Mihalcea. 2020
71 citations recorded by the source corpus.
Cross-domain sentiment analysis has received significant attention in recent years, prompted by the need to combat the domain gap between different applications that make use of sentiment analysis. In this paper, we take a novel perspective on this task by exploring the role of external commonsense knowledge. We introduce a new framework, KinGDOM, which utilizes the ConceptNet knowledge graph to enrich the semantics of a document by providing both domain-specific and domain-general background concepts. These concepts are learned by training a graph convolutional autoencoder that leverages inter-domain concepts in a domain-invariant manner. Conditioning a popular domain-adversarial baseline method with these learned concepts helps improve its performance over state-of-the-art approaches, demonstrating the efficacy of our proposed framework.
Subfields: Sentiment & Opinion Analysis, Knowledge Graphs & Representation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Knowledge Graphs & Representation
Citations in the corpus
- CausaLM: Causal Model Explanation Through Counterfactual Language Models (2020, background): “The core idea of DANN is to reduce the domain gap, by learning common representations that are indistinguishable to a domain discriminator (Ghosal et al. 2020).”
- Continual Pre-training of Language Models for Math Problem Understanding with Syntax-Aware Memory Network (2022, methodology): “, 2020) utilize both domain-related corpus and new pre-training objectives for continual pretraining, or revise the Transformer structure of PLMs for better adaption (Ghosal et al., 2020).”
- Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectives (2022, unknown): “It has motivated several studies (Li et al., 2019b; Shen et al., 2018; Rocha and Cardoso, 2019; Ghosal et al., 2020).”
- Neural Unsupervised Domain Adaptation in NLP—A Survey (2020, unknown): “, 2018) DSN (GSN) 3 (Rocha and Lopes Cardoso, 2019)* DANN, Shared encoders 3 3 (Ghosal et al., 2020) DANN (concept embeddings) 3 (Naik and Rose, 2020) DANN (context embeddings) 3 Data-centric: (Ruder and Plank, 2018) SSL, Multitask tri-training 3 3 (Lim et al.”