Relationship extraction method based on entity mapping

The invention belongs to the field of text information processing, and discloses a relation extraction method based on entity mapping, which comprises the following steps of: firstly, acquiring original data, marking the data and dividing the data into a training set and a test set; then, constructing a CasRelBLCF network model, including sentence semantic embedding, sentence head entity decoding and sentence tail entity decoding; then, the CasRelBLCF network model is trained, and a relation extraction model is obtained; and finally, inputting sentences into the relation extraction model, and extracting a relation triple. According to the relation extraction method based on entity mapping, the probability of illegal label generation can be reduced, the problem of sample imbalance in a tail entity decoder is effectively solved, and the extraction accuracy and the recall rate are effectively improved. 本发明属于文本信息处理领域，公开了一种基于实体映射的关系抽取方法，包括以下步骤：首先，获取原始数据，对数据进行标注后划分为训练集和测试集；然后，构建CasRelBLCF网络模型，包括句子语义嵌入、句子头实体解码、句子尾实体