As the double-carbon concepts of “carbon neutralization” and “carbon peak” are gradually recognized by more and more people, more knowledge in the field of double carbon needs to make people understand it more systematically, to better understand the meaning of double-carbon better. For this reason, this paper proposes a method to construct a knowledge graph in the double carbon field. Firstly, aiming at the semi-structured and unstructured data in the double carbon field, the user-defined Web data processor and MYOMODE parsing method are used to extract the triple. Secondly, through the double carbon-related knowledge obtained from the massive Internet data, the knowledge will be converted into interrelated ternary data. The experimental results show that this method can quickly get valuable and coresponding double-carbon domain knowledge from massive Internet data, and construct it as a knowledge graph.