Construction And Application Of Knowledge Graph Of Crop Growth And Development

Mastering knowledge of the crop production process,including photosynthesis,nutritional requirements,pests and diseases,and production measures,plays an important role in optimizing crop yield and quality,reducing production costs,increasing economic efficiency,and protecting the environment.However,knowledge of crop growth and development is widely dispersed among various channels,including agricultural production practices,online resources,agricultural scientific databases,and scientific monographs.The content involved is extensive,and knowledge resources are in a relatively disordered and scattered state.Building a knowledge graph of crop growth and development can integrate multiple heterogeneous data and knowledge sources to help users understand the rules and relationships of crop growth and development,manage and regulate the crop growth and development process,improve efficiency,reduce resource waste and environmental pollution,and promote sustainable development in agriculture.Therefore,starting from the collection and processing of crop growth and developmentrelated knowledge,this thesis collects crop growth and development-related data resources from different sources and types,obtains,stores,and extracts knowledge from these different data sources,studies the key technologies and methods for constructing crop growth knowledge graphs,explores the application of knowledge graphs in crop growth and development question answering,and verifies whether the constructed knowledge graph can meet the needs of integrating multiple heterogeneous data and knowledge in crop growth and development.This thesis analyzes the origin and current status of research on crop field models and,based on exploring the origin and development status of knowledge graphs,studies the technical progress of combining crop models with knowledge graphs.The thesis reviews the key concepts and construction technologies of knowledge graphs,including ontology model construction,knowledge extraction,knowledge fusion,knowledge storage,and applicationrelated theoretical knowledge of knowledge graphs.Following the principle of ontology model construction,this thesis constructs an ontology model for crop growth and development field using knowledge sources such as scientific monographs in the crop field and agricultural science thesaurus,including 20 core classes,11 object properties,and 17 data properties.Based on the ontology model and data sources,this thesis constructs the data layer of the crop growth and development knowledge graph.By manually annotating the contents of five scientific monographs in the crop field,with a total of 410,000 words,a dataset is constructed for training named entity recognition models.This thesis studies the use of different sources and structured data,including Scrapy framework,BERT-Bi LSTMCRF deep learning model,and other technologies for knowledge extraction and fusion,and constructs the crop growth and development knowledge graph through Neo4 j graph database,including 55,199 different entities,33,520 entity relationships,and 92,393 entity properties.Based on this,utilizing the constructed crop growth and development knowledge graph,a crop growth and development knowledge graph question-answering system was built under the Django framework,which includes graph visualization and knowledge querying functionalities.The test results of the system demonstrate that it is capable of graph visualization and accurate answering of queries,thereby confirming the rationality of the crop growth and development ontology model and knowledge graph construction.