NRE icon indicating copy to clipboard operation
NRE copied to clipboard

Neural Relation Extraction, including CNN, PCNN, CNN+ATT, PCNN+ATT

Results 16 NRE issues
Sort by recently updated
recently updated
newest added

您论文里提到训练集有522611句子、测试集有172448句子。但在您发布的data.zip文件中测试集行数为172448,但句子去重后为61707;训练集行数为570088,句子去重后为368099,即使句子+实体对+关系联合再去重后也是510415,而非522611。 请问是哪里出了问题?您论文中的“句子数量”指的是什么?

Would you like to share how the word embedding file was created, like what procedure was used. And also if I want this algorithm to work on my dataset, how...

Hi, there are two questions bother me a lot, first does the **r** already exist or we need to train it? second in my view, this method can not extract...

我通过 http://iesl.cs.umass.edu/riedel/ecml/ 直接解析出的实体ID都是以/guid/开头,而NRE中处理出的实体id是以m.xxx开头,请问这两个id的对应关系是怎么得到的?

This part in CNN+ATT is not consistent with CNN+ONE, PCNN+ONE and PCNN+ATT.