Many Artificial Intelligence tasks need large amounts of commonsense knowledge. Because obtaining this knowledge through machine learning would require a huge amount of data, a better alternative is to elicit it from people through human computation. We consider the sentiment classification task, where knowledge about the contexts that impact word polarities is crucial, but hard to acquire from data. We describe a novel task design that allows us to crowdsource this knowledge through Amazon Mechanical Turk with high quality. We show that the commonsense knowledge acquired in this way dramatically improves the performance of established sentiment classification methods.
Aspect-based sentiment analysis aims to identify the sentiment polarity of a specific aspect in product reviews. We notice that about 30% of reviews do not contain obvious opinion words, but still convey clear human-aware sentiment orientation, which is known as implicit sentiment. However, recent neural network-based approaches paid little attention to implicit sentiment entailed in the reviews. To overcome this issue, we adopt Supervised Contrastive Pre-training on large-scale sentiment-annotated corpora retrieved from in-domain language resources. By aligning the representation of implicit sentiment expressions to those with the same sentiment label, the pre-training process leads to better capture of both implicit and explicit sentiment orientation towards aspects in reviews. Experimental results show that our method achieves state-of-the-art performance on SemEval2014 benchmarks, and comprehensive analysis validates its effectiveness on learning implicit sentiment.
Aspect-based sentiment analysis (ABSA) is an emerging fine-grained sentiment analysis task that aims to extract aspects, classify corresponding sentiment polarities and find opinions as the causes of sentiment. The latest research tends to solve the ABSA task in a unified way with end-to-end frameworks. Yet, these frameworks get fine-tuned from downstream tasks without any task-adaptive modification. Specifically, they do not use task-related knowledge well or explicitly model relations between aspect and opinion terms, hindering them from better performance. In this paper, we propose SentiPrompt to use sentiment knowledge enhanced prompts to tune the language model in the unified framework. We inject sentiment knowledge regarding aspects, opinions, and polarities into prompt and explicitly model term relations via constructing consistency and polarity judgment templates from the ground truth triplets. Experimental results demonstrate that our approach can outperform strong baselines on Triplet Extraction, Pair Extraction, and Aspect Term Extraction with Sentiment Classification by a notable margin.
Cambria, Erik (Nanyang Technological University) | Poria, Soujanya (Nanyang Technological University) | Hazarika, Devamanyu (National University of Singapore) | Kwok, Kenneth (Institute of High Performance Computing, A*STAR)
With the recent development of deep learning, research in AI has gained new vigor and prominence. While machine learning has succeeded in revitalizing many research fields, such as computer vision, speech recognition, and medical diagnosis, we are yet to witness impressive progress in natural language understanding. One of the reasons behind this unmatched expectation is that, while a bottom-up approach is feasible for pattern recognition, reasoning and understanding often require a top-down approach. In this work, we couple sub-symbolic and symbolic AI to automatically discover conceptual primitives from text and link them to commonsense concepts and named entities in a new three-level knowledge representation for sentiment analysis. In particular, we employ recurrent neural networks to infer primitives by lexical substitution and use them for grounding common and commonsense knowledge by means of multi-dimensional scaling.
Fu, Peng (Institute of Information Engineering, Chinese Academic of Sciences) | Lin, Zheng (Institute of Information Engineering, Chinese Academic of Sciences) | Yuan, Fengcheng (Institute of Information Engineering, Chinese Academic of Sciences) | Wang, Weiping (Institute of Information Engineering, Chinese Academic of Sciences) | Meng, Dan (Institute of Information Engineering, Chinese Academic of Sciences)
Context-based word embedding learning approaches can model rich semantic and syntactic information. However, it is problematic for sentiment analysis because the words with similar contexts but opposite sentiment polarities, such as good and bad, are mapped into close word vectors in the embedding space. Recently, some sentiment embedding learning methods have been proposed, but most of them are designed to work well on sentence-level texts. Directly applying those models to document-level texts often leads to unsatisfied results. To address this issue, we present a sentiment-specific word embedding learning architecture that utilizes local context informationas well as global sentiment representation. The architecture is applicable for both sentence-level and document-level texts. We take global sentiment representation as a simple average of word embeddings in the text, and use a corruption strategy as a sentiment-dependent regularization. Extensive experiments conducted on several benchmark datasets demonstrate that the proposed architecture outperforms the state-of-the-art methods for sentiment classification.