Modelpublic

manmanluo/amazon-iphone8plus-sentiment-bert

sourceHugging Faceupdated 9mo agoView on Hugging Face

0likes6downloads

Model Card

BERT-based Sentiment Classifier (Amazon iPhone 8 Plus)

Model description

This repository contains a fine-tuned BERT model (bert-base-uncased) for binary sentiment classification (positive vs. negative) on Amazon iPhone 8 Plus reviews.

The model is fine-tuned using PyTorch and the Hugging Face Transformers library.

Dataset

The model is trained and evaluated on a curated subset of the Amazon Reviews 2023 dataset (McAuley Lab), restricted to the iPhone 8 Plus product family.

Dataset repository:

—https://huggingface.co/datasets/manmanluo/amazon-iphone8plus-review

Evaluation

Train/validation/test splits are shared across models for controlled comparison. Evaluation results and training details are provided in:

—results/bert/training_summary.json
—results/bert/test_metrics.json

The decision threshold is fixed at 0.5.

Intended use

This model is intended for research and baseline comparison purposes only. It is not optimized for production use.