京东宠物用品评价中英双语平行语料集

本数据集是基于大语言模型虚拟生成的中英双语宠物用品评价平行语料集，共100,000条有效数据。数据模拟京东电商平台的宠物用品评价场景，每条数据均包含中文评价文本和英文翻译版本，全面覆盖宠物饮食、窝垫起居、清洁洗护、出行牵引、玩具磨牙、如厕尿砂、服饰配饰七大品类共70款产品，每条评价标注正面/中性/负面情感极性以及1-5级情感强度。数据集面向NLP研究者开放，适用于情感分析、文本分类、机器翻译、评价挖掘、数据增强等多项任务。This dataset is a parallel corpus of Chinese-English pet product reviews virtually generated using Large Language Models (LLMs), consisting of 100,000 valid entries. The data simulates the pet product review scenario on the JD.com e-commerce platform. Each entry contains both Chinese review text and its corresponding English translation, covering 70 products across 7 major categories: pet diet, bedding & living supplies, cleaning & grooming products, travel & traction gear, chew toys, toilet & litter supplies, and apparel & accessories. Each review is annotated with sentiment polarity (positive/neutral/negative) and a sentiment intensity level ranging from 1 to 5. This dataset is open to NLP researchers and supports multiple tasks including sentiment analysis, text classification, machine translation, review mining, and data augmentation.

搜集汇总

背景与挑战

背景概述

该数据集是一个基于大语言模型虚拟生成的宠物用品评价中英双语平行语料集，共包含10万条数据。它模拟了京东电商平台的评价场景，覆盖七大产品品类，每条数据均标注了情感极性及强度，适用于情感分析、机器翻译等多种自然语言处理任务。