In the era of big data, efficiently managing and leveraging extensive data is crucial, with knowledge graph being pivotal for intelligent information processing. Ontologies, as the core of knowledge graph, provide a structured model for knowledge representation and semantic relationships. However, traditional static ontologies encounter difficulties in keeping up with domain knowledge updates and adapting to new queries, which hampers their ability to meet personalized information needs. This paper addresses these challenges by proposing a dynamic ontology framework. It aims to design and implement a knowledge graph construction tool that can adapt to changes. The tool allows for agile updates at the physical layer and flexible application at the logical layer. Additionally, it includes a visual operation tool that makes handling complex ontologies and knowledge graphs accessible to non-experts, bridging the gap between research and practical application.