Research On Measuring Instrument Information Platform Construction Technology And Intelligent Recommendation Algorithm

Metrology is the science of measuring.With the continuous development of society,the measurement work has penetrated into the fields of industry,agriculture,commerce and high-precision aerospace,artificial intelligence,etc.,What followed was the continuous growth of the measurement business and measurement field constant expansion.Although the Institute of Metrology of Shaanxi Province has its own information platform,it is in the semi-information and semi-manualization stage.The functions are scattered,the system's software and hardware platforms are not uniform,and new modules are needed in the business development.Therefore,a measurement management platform with a higher degree of informationization and more complete functions is needed.In addition,for the continuous development of information platform,the research of recommended algorithms is also essential.This paper is mainly adout the construction of the measurement information platform and the recommendation algorithm.For the construction of measuring instrument information platform combined with existing software and hardware resources and user needs analysis,using B/S structure system for design,Develop with ASP.Net technology on the.Net platform,the overall development uses MVC design pattern to achieve front and rear Separate development,make business logic more clearer.The front end uses MiniUI,bootstrap,the background API performs data storage and extraction by interacting with the database.The main components of the system are instrument transceiver module,certificate making module,quality management module,printing charging module,strong inspection appliance supervision module and warehouse management module,which contain all the services required for the measurement information platform.At present,the information platform has been officially put into use,which proves that the platform can meet the functions required by customers while running smoothly,realizes a fully informationized office mode,and improves office efficiency.For the recommendation algorithm,mainly researched on collaborative filtering algorithm,Weighted slope one algorithm and Collaborative filtering algorithm for improving predictive scoring matrix.Traditional collaborative filtering algorithms have problems such as sparse data and poor scalability.The weighted slope one algorithm based on linear regression analysis and adds a common scoring cardinality to the traditional slope one algorithm,which enhances the accuracy of the algorithm and has strong scalability.Collaborative filtering algorithm for improving predictive scoring matrix,filling a sparse scoring matrix with a weighted slope one algorithm,Then combined with collaborative filtering algorithm for recommendation.The experimental results show that the collaborative filtering algorithm using improved predictive scoring matrix is the most accurate in several algorithms,which reduces the recommendation error.