The Construction Of Plant Knowledge Graph Based On Deep Learning

With the rapid development of information technology,especially the wide application of artificial intelligence technology,how to process the massive data in the Internet and provide web users with easy-to-understand knowledge has become a research hotspot.Knowledge graph plays an important role in transforming data into knowledge.knowledge graph represents the entities,concepts and their relations in the real world.In the field of forestry,knowledge graph has a wide range of applications in semantic retrieval,data mining,predictive analysis,intelligent question answering,and decision-making.In the field of plant information,this paper research on the construction of plant knowledge graph based on multiple data sources.We work on the construction of knowledge graph,named entity recognition and semantic retrieval based on knowledge graph.The process of constructing the domain knowledge graph is analyzed,and the important technologies in the process are expounded.We summarize the characteristics of texts in flora of China,and study how to recognize attribute named entities in the texts.Then we propose a method based on deep neural network for named entity recognition.Several data sources which related to plant information are researched,including forestry websites and encyclopedia websites.The constructed plant knowledge graph is applied to the semantic retrieval problem,and a semantic retrieval system based on plant knowledge graph is constructed by using visualization technology.The plant knowledge graph is an attempt to construct the knowledge graph in the field of forestry,which can provide users with intelligent and intuitive semantic retrieval.Furthermore,we can integrate other forestry knowledge into plant knowledge graph.Thus,the plant knowledge graph can be a knowledge support for the development of smarter forestry,which can serve for semantic retrieval,knowledge reasoning and decision-making in forestry.