Sentiment Analysis of E-Commerce Food Reviews Considering Explainability and Efficiency: An AdapterFusion-Based Approach

View at Publisher

Ask AI

Bookmark

Share

Key Points

This research aims to improve sentiment analysis of e-commerce food reviews by addressing automatic labeling limitations.
Proposed an AdapterFusion-based sentiment analysis framework to analyze consumer reviews.
Developed human-validated silver labels by combining LLM-assisted labeling and researcher verification.
Quantified adapter contributions by attribute for model interpretability and extensibility.
Achieved performance comparable to full parameter fine-tuning using only 2-3% of parameters, enhancing labeling reliability.
Enhanced interpretability of model decisions through quantified attribute contributions.

Answers are grounded in this paper and cite their sources.

Demonstrates improved sentiment analysis efficiency in e-commerce reviews, indicating enhanced customer insights.

No abstract yet. The journal hasn't published one we can show.

No takes yet. Share an insight, caveat, or question.

Choi et al. (2026) studied this question.

View at Publisher

Ask AI

Bookmark

Share