sukhrobnurali/uzbek_spam_dataset
收藏Hugging Face2025-12-18 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/sukhrobnurali/uzbek_spam_dataset
下载链接
链接失效反馈官方服务：
资源简介：
这是一个用于训练乌兹别克语垃圾信息检测模型的合成数据集，特别针对Telegram风格的消息。数据集包含2000条乌兹别克语文本消息，标记为spam或normal。数据集涵盖了多种垃圾信息类别（如激进广告、快速致富计划、虚假工作机会等）和正常消息类别（如日常问候、工作消息、家庭对话等）。数据采用拉丁和西里尔字母的乌兹别克语，分为1800条训练数据和200条测试数据。
A synthetic dataset for training spam detection models for Uzbek language, specifically targeting Telegram-style messages. The dataset contains 2000 Uzbek text messages labeled as either spam or normal, covering various spam categories (such as aggressive advertising, get-rich-quick schemes, fake job offers) and normal message categories (like casual greetings, work messages, family conversations). The data uses Uzbek in both Latin and Cyrillic scripts, with 1800 samples for training and 200 for testing.
提供机构：
sukhrobnurali
