paper-with-me

Papers

Position-based Contributive Embeddings for Aspect-Based Sentiment Analysis

2021-08-11 · Zijian Zhang, Chenxin Zhang, Jiangfeng Li, Qinpei Zhao

Aspect-based sentiment analysis (ABSA), exploring sentiment polarity of aspect-given sentence, is a fine-grained task in the field of nature language processing. Previously researches typically tend to predict polarity based on the meaning of aspect and opinions. However, those approaches mainly focus on considering relations implicitly at the word level, ignore the historical impact of other positional words when the aspect appears in a certain position. Therefore, we propose a Position-based Contributive Embeddings (PosCE) to highlight the historical reference to special position aspect. Contribution of each positional words to the polarity is similar to the process of fairly distributing gains to several actors working in coalition (game theory). Therefore, we quote from the method of Shapley Value and finally gain PosCE to enhance the aspect-based representation for ABSA task. Furthermore, the PosCE can also be used for improving performances on multimodal ABSA task. Extensive experiments on both text and text-audio level using SemEval dataset show that the mainstream models advance performance in accuracy and F1 (increase 2.82% and 4.21% on average respectively) by applying our PosCE.

📄 PDF Abstract BibTeX arXiv:2108.05098

Code (0)

등록된 구현이 없습니다.

Tasks

Aspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)PositionSentenceSentiment Analysis

Similar Papers 제목 키워드 기반

An Aspect Extraction Framework using Different Embedding Types, Learning Models, and Dependency Structure

2025-03-05 · Ali Erkan, Tunga Güngör

Aspect-based sentiment analysis has gained significant attention in recent years due to its ability to provide fine-grained insights for sentiment expressions related to specific features of entities. An important compon…

Aspect-Based Sentiment AnalysisAspect ExtractionDependency ParsingSentiment Analysis

Hierarchical Attention Based Position-Aware Network for Aspect-Level Sentiment Analysis

2018-10-01 · CONLL 2018 10 · Lishuang Li, Yang Liu, AnQiao Zhou

Aspect-level sentiment analysis aims to identify the sentiment of a specific target in its context. Previous works have proved that the interactions between aspects and the contexts are important. On this basis, we also …

Aspect-Based Sentiment Analysis (ABSA)Feature EngineeringPositionSentence+1

Label Correction Model for Aspect-based Sentiment Analysis

2020-12-01 · COLING 2020 8 · Qianlong Wang, Jiangtao Ren

Aspect-based sentiment analysis includes opinion aspect extraction and aspect sentiment classification. Researchers have attempted to discover the relationship between these two sub-tasks and have proposed the joint mode…

Aspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)Aspect Extractionmodel+3

A Hybrid Approach for Aspect-Based Sentiment Analysis Using Deep Contextual Word Embeddings and Hierarchical Attention

2020-04-18 · Maria Mihaela Trusca, Daan Wassenberg, Flavius Frasincar, Rommert Dekker

The Web has become the main platform where people express their opinions about entities of interest and their associated aspects. Aspect-Based Sentiment Analysis (ABSA) aims to automatically compute the sentiment towards…

Aspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)Sentiment AnalysisWord Embeddings

Modelling Context and Syntactical Features for Aspect-based Sentiment Analysis

2020-07-01 · ACL 2020 6 · Minh Hieu Phan, Philip O. Ogunbona

The aspect-based sentiment analysis (ABSA) consists of two conceptual tasks, namely an aspect extraction and an aspect sentiment classification. Rather than considering the tasks separately, we build an end-to-end ABSA s…

Aspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)Aspect ExtractionSentence+2