基于特征向量的煤矿领域实体关系抽取

Extraction of Entity Relationship in Coal Mine Based on Feature Vector

煤炭是我国的基础能源,煤矿安全是煤矿生产的重中之重。一个煤矿领域的知识图谱可以很好地帮助煤矿工作人员分析常见煤矿事故。知识图谱是一种揭示实体之间关系的语义网络,通过对原始数据进行关系抽取、质量评估,从而对现有知识图谱进行更新。通过引入因子图模型,基于特征向量对煤矿领域实体间的发生关系进行抽取,并结合远程监督方法及启发式规则降低实体关系抽取的人工标注量及噪声比率。实验结果表明,上述方法对于煤矿领域"发生"关系的抽取具有较高的准确率。

Coal is the basic energy of our country,coal mine safety is the most important part of coal production.A knowledge map of the coal field can help the coal mine workers to analyze the common coal mine accidents.Knowledge graph is a semantic network that reveals the relationship between entities.It updates the existing knowledge graph by extracting the relationship from the original data and evaluating its quality.Introduces a factor graph model to extract the occurrence relationship between entities in the coal mine field based on eigenvectors, and remote monitoring method and heuristic rules are combined to reduce the manual annotation and noise ratio of entity relationship extraction.The experimental results show that this method has a high accuracy in extracting the occurrence relationship in the field of coal mine.