amazon-reviews

# Dataset Card for Amazon Reviews 2018

This dataset is a collection of title-review pairs collected from Amazon, as collected in [Ni et al.](https://cseweb.ucsd.edu//~jmcauley/pdfs/emnlp19a.pdf). See [Amazon Reviews 2018](https://nijianmo.github.io/amazon/index.html) for additional information.
This dataset can be used directly with Sentence Transformers to train embedding models.

## Dataset Subsets

### `pair` subset

* Columns: "title", "review"
* Column types: `str`, `str`
* Examples:
```python
{
'title': "It doesn't fit my machine. I can't seem to ...",
'review': "It doesn't fit my machine. I can't seem to find the part for the Mr. Coffee Tea maker I have.",
}
```
* Collection strategy: Reading the Amazon Reviews 2018 dataset from [embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data).
* Deduplified: No

# 亚马逊评论2018数据集卡片

本数据集为从亚马逊平台采集的标题-评论对集合，相关采集工作见于[Ni等人](https://cseweb.ucsd.edu/~jmcauley/pdfs/emnlp19a.pdf)的研究。更多详细信息可查阅[亚马逊评论2018](https://nijianmo.github.io/amazon/index.html)官方页面。

本数据集可直接配合句子变换器（Sentence Transformers）用于嵌入模型的训练。

## 数据集子集

### `pair` 子集

* 列名："title", "review"
* 列数据类型：`str`, `str`
* 示例：
python
{
'title': "It doesn't fit my machine. I can't seem to ...",
'review': "It doesn't fit my machine. I can't seem to find the part for the Mr. Coffee Tea maker I have.",
}

* 采集策略：从[embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data)数据集仓库中读取亚马逊评论2018数据集。
* 去重状态：未执行去重

搜集汇总

背景与挑战

背景概述

该数据集是2018年亚马逊标题-评论对的集合，主要用于训练句子嵌入模型。其中'pair'子集包含'title'和'review'两列字符串数据，未进行去重处理。