Hyperbolic Disentangled Representation for Fine-Grained Aspect Extraction

Abstract

Automatic identification of salient aspects from user reviews is especially useful for opinion analysis. There has been significant progress in utilizing weakly supervised approaches, which require only a small set of seed words for training aspect classifiers. However, there is always room for improvement. First, no weakly supervised approaches fully utilize latent hierarchies between words. Second, each seed word’s representation should have different latent semantics and be distinct when it represents a different aspect. In this paper we propose HDAE, a hyperbolic disentangled aspect extractor in which a hyperbolic aspect classifier captures words’ latent hierarchies, and an aspect-disentangled representation models the distinct latent semantics of each seed word. Compared to previous baselines, HDAE achieves average F1 performance gains of 18.2% and 24.1% on Amazon product review and restaurant review datasets, respectively. In addition, the embedding visualization experience demonstrates that HDAE is a more effective approach to leveraging seed words. An ablation study and a case study further attest the effectiveness of the proposed components.

Enhanced Machine Reading Comprehension Method for Aspect Sentiment Quadruplet Extraction

“…Aspect Sentiment Classification (ASC) [19,26,4] is to predict the sentiment polarity for a specific aspect within a sentence. More recently, to understand more complete aspect-level opinion and recognize the correspondence and dependency among different sentiment elements, several recent studies have explored the joint detection of multiple sentiment elements in the pair, triplet, or quadruplet formats, including Aspect-Opinion Pair Extraction (AOPE) [11,41], Aspect Sentiment Triplet Extraction (ASTE) [5,31,27], and Aspect Sentiment Quadruplet Extraction (ASQE) [1,3,38]. Compared to the other ABSA tasks, ASQE is considered as the most comprehensive and challenging task due to its complexity in capturing and combining more sentiment elements.…”

“…(ASTE) [5,31,27,33] attempts to extract aspect-opinion-sentiment triplets from the given sentence. Despite the great progress in ABSA, most existing studies have only considered the partial extraction of sentiment elements instead of providing a comprehensive aspect-level sentiment structure.…”