File size: 517 Bytes
38136ed be504c1 38136ed be504c1 38136ed | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | ---
title: ACSA Clothing Sentiment
emoji: "👕"
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 5.49.1
python_version: "3.10"
app_file: app.py
pinned: false
---
# Aspect-Level Sentiment Analysis for Clothing Reviews
Interactive demo and research dashboard for a BERT metadata-fusion ACSA model trained on 30,000 clothing reviews.
## Included inference artifacts
- Proposed metadata-fusion checkpoint and tokenizer
- Matching `meta_encoder.pkl`
- Overall, aspect-level, and ablation evaluation reports |