mashey/dv-synthetic-errors
收藏官方服务：
资源简介：
迪维希语文本错误纠正数据集，包含正确句子和合成生成的错误句子。该数据集旨在测试迪维希语错误纠正模型和工具。数据集结构为输入输出对：correct字段为原始正确句子，incorrect字段为带有合成错误的句子。数据集创建基于迪维希语文章收集，通过字符和变音符号替换生成错误，错误率为每单词30%概率。
Dhivehi text error correction dataset containing correct sentences and synthetically generated errors. The dataset aims to test Dhivehi language error correction models and tools. The dataset structure consists of input-output pairs: correct for original correct sentences and incorrect for sentences with synthetic errors. It was created using a collection of Dhivehi articles with error generation via character and diacritic substitutions at a 30% probability per word.
提供机构：
mashey
