钦杭成矿带斑岩铜矿知识图谱构建及应用展望

知识图谱使用人与机器能共同理解的语言,以"图"的方式来描述真实世界,是人工智能研究的重要方向之一.本研究是构建单体矿床、成矿系列和重要成矿区(带)的知识图谱实验的一部分,收集了钦杭成矿带6个较为典型的斑岩铜矿、斑岩-夕卡岩型铜矿的原始文本数据,参照斑岩铜矿床概念模型进行知识获取,标注、抽提文本中的实体、关系、属性,构建...

Constructing knowledge graph for the porphyry copper deposit in the Qingzhou-Hangzhou Bay area: Insight into knowledge graph based mineral resource prediction and evaluation

Knowledge graphs, fundamental to artificial intelligence, describe the real world in graphic forms using a language that can be understood by both humans and machines. This paper presents a case study on the construction of knowledge graph for porphyry copper deposit. The raw text data were collected and integrated from six selected porphyry and porphyry-skarn copper deposits in the Qinzhou-Hangzhou Bay metallogenic belt, one of the key metallogenic belts of China. The entities, relations and attributes in the text are labeled and extracted in reference to the conceptual model of porphyry copper deposit. The resulted knowledge graph has the basic application functions. As part of a planned integrated knowledge graph—from a single deposit, through upper-geared metallogenic series, to top metallogenic province (belt)—the present study may be extended toward understanding and improving future way of mineral resource prediction and evaluation. The interrelationship among the earth system, the metallogenic system, the exploration system, and the prediction and evaluation system (ES-MS-ES-PS) should be fully understood, and a knowledge graph for the ES-MS-ES-PS system is essential. The key scientific and technological challenges to attain such a large-scale knowledge graph for the ES-MS-ES-PS system thus include system of progressive association of domain ontology and knowledge graph, automation technology for constructing large-scale domain ontology and knowledge graph, self-evolution and complementary techniques for embedding multi-modal correlation data into knowledge graph, and ES-resource prediction theory and methods based on knowledge graph, big-data mining and artificial intelligence.