Design And Implementation Of Product Indicator Prediction System Based On Knowledge Graph

With the development and progress of science and technology and industry,the complexity of functions,configurations and indicators of products in the field of equipment manufacturing is increasing.Product design for customer needs is facing the challenges of increasing personalized demand,changing demand quickly and shorter product design cycle.To solve these problems,this paper designs a product index prediction system based on knowledge graph.On the basis of constructing the knowledge graph of complex product index,this paper uses the inference technology based on knowledge graph to discover the implicit relationship between complex products and indicators,so as to complete the task of predicting the missing index data of the known products and generating a new product plan according to the needs of users.This system can quickly respond to the needs of users and shorten the product design cycle,thus providing technical support for the analysis and design process of complex products.The main function modules of this system are product data maintenance module,product model management module,indicator association prediction module,product plan generation module and program comparison evaluation module.Among them,product data maintenance module is to add,delete,modify and query product categories and products.The product model management module trains different types of models based on user configuration information.The indicator association prediction module predicts the missing indicator data of known products.The product plan generation module generates new product plans that do not exist in the current database based on some configuration information and indicator data,and predicts missing configuration information and indicator data for new product plans.The scheme comparison evaluation module compares the index data of the new product scheme with the known product,and then modifies the index data of the unreasonable new product scheme.This system uses Python language to train different types of models.This system uses Spring Boot and My Batis-Plus framework to implement the system back-end module.This system uses Vue framework and Element UI front-end component library to display front-end page data.This system uses Neo4 j diagram database to store product index data.This system uses Mysql database to store related configuration information.