arXiv:1805.09648
Crowd-Labeling Fashion Reviews with Quality Control
원문 초록
We present a new methodology for high-quality labeling in the fashion domain with crowd workers instead of experts. We focus on the Aspect-Based Sentiment Analysis task. Our methods filter out inaccurate input from crowd workers but we preserve different worker labeling to capture the inherent high variability of the opinions. We demonstrate the quality of labeled data based on Facebook's FastText framework as a baseline.
함께 살펴볼 논문
공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.
M-ABSA: A Multilingual Dataset for Aspect-Based Sentiment Analysis
Multi-View Attention Syntactic Enhanced Graph Convolutional Network for Aspect-based Sentiment Analysis
DS$^2$-ABSA: Dual-Stream Data Synthesis with Label Refinement for Few-Shot Aspect-Based Sentiment Analysis
Evaluating Zero-Shot Multilingual Aspect-Based Sentiment Analysis with Large Language Models
Multi-Domain ABSA Conversation Dataset Generation via LLMs for Real-World Evaluation and Model Comparison
CrosGrpsABS: Cross-Attention over Syntactic and Semantic Graphs for Aspect-Based Sentiment Analysis in a Low-Resource Language
토론
이 논문과 연결된 포럼 글이 아직 없습니다.