E-ABSA20K: A Dataset and Propose-and-Verify for Aspect-Based Sentiment Analysis in Long E-commerce Reviews

Aspect-Based Sentiment Analysis (ABSA) is critical for extracting actionable product insights from e-commerce reviews. However, most public ABSA benchmarks are restricted to short texts and a limited range of domains, and therefore underrepresent the challenges posed by real-world reviews-where multiple aspects co-occur, colloquial and noisy expressions are common, and evidence must often be aggregated across sentences in long contexts.

We introduce E-ABSA20K, a multi-domain dataset of 20K reviews from four product categories (WomenBags, Dresses, Cosmetics, and Furniture), annotated with review-level sentiment quads. Compared to existing benchmarks, E-ABSA20K contains substantially longer and more aspect-dense reviews, averaging 63.9 words and 6.0 quads per review. We further propose a two-stage propose-and-verify framework for review-level quad extraction (target, aspect, opinion, sentiment). The first stage generates high-recall candidates under strict schema constraints, while the second stage conducts explicit grounding, scope, and modality verification, followed by review-level consolidation to mitigate hallucinations and scope leakage in long reviews. Experiments across multiple Qwen3 model sizes demonstrate that our approach consistently outperforms singlestage prompting (with and without chain-ofthought) as well as competitive ABSA extraction baselines, improving quad-level micro-F1 (µF) and robustness on discourse-hard cases such as comparisons and conditionals. The dataset was collected from a major Englishlanguage e-commerce platform in Southeast Asia; while reviews are predominantly written in English, a small fraction (approximately 0.8%) contains code-mixed tokens, reflecting realistic multilingual user-generated content.