Representation and Association of Chinese Financial Equity Knowledge Driven by Multilayer Ontology

EasyChair Preprint 5262

Abstract

Aiming at the current situation of complex financial ownership structure and isolated data organization, this study referring to the methods for multi-layer hierarchical construct domain ontology modeling. At the same time, the three dimensions of industry, company and internal environment were integrated, and the concept cube was designed and constructed based on knowledge extraction and text classification technology, so as to provide a multi-level and fine-grained knowledge representation and association method for financial equity knowledge. The experimental results show that conceptual cube structure represents semantic information as a dense low-dimensional representation vector, which greatly enhances semantic relevance and interpretability. The multi-layer ontology-driven ownership structure reflects a variety of knowledge association patterns, and in the "Intelligent Financial Big Data System" developed by the research team, the association query of three categories of association relationships in the field of industry, enterprise and internal environment is realized, as well as the dynamic analysis and supervision of typical financial management problems.

Keyphrases: Concept cube, Financial equity, Multilayer domain ontology, Triple Extraction, knowledge association, knowledge representation

@booklet{EasyChair:5262,
  author    = {Zhenghao Liu and Huakui Lv and Xi Zeng},
  title     = {Representation and Association of Chinese Financial Equity Knowledge Driven by Multilayer Ontology},
  howpublished = {EasyChair Preprint 5262},
  year      = {EasyChair, 2021}}