A Healthy Diet Knowledge Q/A System Based On Knowledge Graph

The World Health Organization has proposed that an unhealthy diet is a major risk factor for a series of chronic diseases such as cardiovascular disease,cancer,diabetes,and other obesity-related problems.A healthy diet can not only reduce the incidence of certain diseases,but also can adjust the symptoms to improve the physical condition after the disease;in recent years,the public has increasingly paid more attention to healthy and reasonable diets and tried to search for vegetarian Get dietary messages that meet your needs on the Search Engine,or obtain healthy dietary knowledge by pushing messages from various unknown sources.Based on this reason,The 19 th National Congress of the Communist Party of China has therefore made a major decision to implement the "Healthy China" strategy,raising the maintenance of people's health to the height of the national strategy.Relevant government departments provide health diet knowledge information services to the public is an important measure to implement the "Healthy China" strategy.A scientific and effective intelligent Q/A system is one of the more advanced information service models.Traditional Q/A system returns keywords by matching keywords.Which cannot parse the user's semantic information and not logical,so they cannot meet the user's needs.Knowledge Graph provides a method to analyze things from the perspective of "relationship".The Healthy Diet Knowledge Q/A system based on knowledge Graph can analyze the semantics of the original questions entered by the users,and accurately and quickly use human-friendly Ways to give users the answers and related knowledge they need.The article uses the Python Scrapy crawler framework to crawl the relevant data of healthy meals on Baidu Encyclopedia website.The crawled data and the contents of ?Famous Old Chinese Doctor Li Qiangou's Diet Secret Recipe-Food Volume? are used to build healthy diet knowledge graph.Then,named entity recognition and relation extraction were performed by CRF ++ tool and the Deepdive relationship extraction project.Finally,a knowledge graph in the field of healthy diet was successfully constructed using Neo4 j graph database.The Healthy Diet knowledge graph including 1294 nodes and 2069 triples.At the same time,based on the healthy diet knowledge graph,a healthy diet knowledge intelligent Q/A system was developed.First,the system design and technical route analysis of the healthy diet knowledge Q/A system were performed.Then,HanLP word segmentation was used to perform word segmentation and key feature extraction on the original questions raised by users.Use the Na?ve Bayesian Classification based on Spark to classify and match the original question.Based on the above results,go to the graph database Neo4 j to find the answer.The intelligent dietary Q/A system for healthy dietary knowledge developed by the article uses Spring-Boot to complete the development of the overall system framework.When the system is successfully researched and developed,the intelligent interactively question answering function can be come true.Finally,the function implementation status of the intelligent dietary Q/A system for healthy dietary knowledge developed was verified through a combination of running demonstration and experimental testing,and the actual use effect of the system was tested.After testing,the system can realize the eight functions preset in the system design phase;the actual use satisfaction rate of the system has reached 65%,and there is still some room for improvement before it is actually put in the market.