Methods of Composing Hierarchical Knowledge Graphs of Telecommunication Network

Abstract

The article is devoted to the issue of developing and analyzing methods of composing hierarchical knowledge graphs of telecommunication networks with a view to representing network data. The article substantiates the timeliness of discussing this issue and describes the classes of problems that can be solved using hierarchical graph models. It discusses the structure of particular hierarchical graph models, their layers, and how it is possible to connect these models layer by layer and create a single hierarchical graph model. It also discusses data formats and approaches to adding dynamic data that are applicable to the resulting hierarchical knowledge graph. It studies the advantages of the hierarchical knowledge graph over one-level analogs. To explain the advantages of using a hierarchical knowledge graph, an example of modeling such a knowledge graph for a cable TV operator’s network is given and compared with a one-level knowledge graph using a specific problem encountered by the operator. It is shown that it is much less difficult to execute queries of the same kind using a hierarchical graph. At the end of the article, conclusions are made and directions for future research are discussed.