xabpuk/yelp_review_full
收藏资源简介：
YelpReviewFull数据集是一个用于文本分类和情感分类的英语评论数据集，包含来自Yelp的评论，数据提取自Yelp Dataset Challenge 2015。数据集由650,000个训练样本和50,000个测试样本组成，每个样本包括评论文本和对应的星级标签（1到5星），其中文本经过转义处理（如双引号转义为两个双引号，换行符转义为 ）。该数据集最初由Xiang Zhang等人构建，并用作文本分类基准，支持多类分类任务，评估指标包括准确率、F1分数、精确率和召回率等。
The Yelp reviews dataset consists of reviews from Yelp. It is extracted from the Yelp Dataset Challenge 2015 data. The dataset is mainly used for text classification and sentiment classification, with reviews written in English. It contains 650,000 training samples and 50,000 testing samples, each comprising a review text and a corresponding star label (from 1 to 5 stars). The text fields are escaped (e.g., double quotes escaped as double double quotes, new lines as ). The dataset was constructed by Xiang Zhang from the Yelp Dataset Challenge 2015 and serves as a benchmark for text classification tasks, supporting multi-class classification with metrics such as accuracy, F1, precision, and recall.
