Research on Knowledge Graph Technology with Unbalanced Samples

Abstract

The application of the existing knowledge graph is widely concentrated in the civil fields with sufficient sample size and uniform distribution, especially in the Internet, finance and other fields. Due to the particularity of its application background in the military field, there is a general lack of sample size, which greatly limits the support ability of knowledge atlas technology for military intelligent applications. Therefore, how to solve the problem of automatic knowledge extraction of multimodal data under the condition of unbalanced samples is the key core problem that affects the application effect of knowledge perception technology in the military field. This topic focuses on the characteristics of small batch, multi batch, imbalance and diversity of data in the field of military operations, introduces the self-monitoring method based on deep learning, solves the problem of low accuracy of automatic knowledge extraction of unstructured data, solves the dependence of traditional intelligent extraction algorithm on sample size, carries out research on automatic construction technology of knowledge atlas, and constructs equipment type spectrum library.