Research And Implementation Of Distributed Wine Information Analysis System

With the development of wine industry in our country,the traditional way of wine enterprise information management mode has been unable to meet the requirements of enterprise,in view of the large amount of data the wine information data processing,and enterprise information resource management level and work efficiency,this paper designed and implemented the wine information analysis system based on distributed.The system effectively realizes the integration of wine information and the standardization and standardization of information management,and improves the efficiency of information management on personnel and wine.Based on the background of the wine industry,data mining and big data cloud computing technology,this system has designed and implemented a distributed wine information analysis system.The system mainly includes seven function modules:foreground information display module,welcome home page module,data import export module,information management module,information analysis prediction module,user management module and data information visualization analysis module.The front-end information display module is mainly responsible for displaying the wine products,recruitment information and contact information of the winery to the users.Welcome home page module is responsible for the display system of profile,such as weather conditions,the current time near chateau,chateau map,sales information,sales area,wine storage and user information,etc.;The data import and export module is mainly responsible for the import and export system of wine data.The information management module is mainly responsible for the management of winery weather information,winery environment information,wine composition information,wine information,inventory information,warehousing information and sales information.Information module is mainly responsible for the chateau within 48 hours of the weather,hot word search as well as the level of wine quality prediction analysis,including the level of wine quality analysis in machine learning algorithms for theoretical support,using the hadoop technology based on distributed to forecast analysis;The user management module is mainly responsible for the management of system member information and the allocation of role authority.The data visualization display module is mainly responsible for the friendly visualization of the data information managed in the system.The main work of this paper is as follows:1.Major technical programs.On the level of wine quality prediction analysis module,the system adopts the way the hadoop technology based on distributed data processing,data preprocessing used ETL data warehouse technology,data visualization using the Echarts technology,data of asynchronous request processing using Ajax technology.2.System design and implementation.In Windows and Linux for development platform,this study based on open source JAVA development language,eclipse3.8 as development tool,for open source relational MySql database,the hadoop technologies and JAVA Web technology based on distributed autonomous wine information analysis system was designed and developed.3.In this paper,the support vector machine(SVM)algorithm,naive bayesian algorithm,BP neural network and deep learning algorithm are studied for the prediction and analysis of wine data grade quality.Experiments,using the UCI lab white wine and red wine as the training sample data set,respectively,using the above machine learning algorithms to forecast the level of wine quality analysis contrast,according to the experimental results,select design has realized the parallel algorithm of SVM based on hadoop.Experiments show that the parallel SVM has a significant improvement in efficiency when training and forecasting large data samples,and the prediction accuracy will not be significantly reduced.