Research On The Price Information Monitoring And Analysis System Of Vegetable&Fruit Based On Big Data

At present,our country agricultural production exist serious problem between production and market,this is mainly by the agricultural policy makers only by speculation or use the decision-making accuracy is not high due to the decision system of decision-making production.With lots of agricultural price for this article is based on large data information as the main body,combined with the weather affect the price of agricultural products,production and other factors affect the prices of agricultural,price forecasting,regional distribution,reasonable distribution,etc are discussed in this paper,aims to find more suitable for the analysis of the large data of agricultural production decision process,use big data to improve the precision of the agricultural production decision as soon as possible.Based on the existing big data technology,this paper designs the flow of price information monitoring and analysis of vegetable and fruit.The application of web crawler tools to climb the agricultural prices,yield and other related data on each agricultural website.The data of agricultural prices in China were analyzed by using big data thinking.To design prototype system,mainly considering the limitations of the traditional technology in front of huge amounts of data,the application of the technology of big data related to overcome the traditional technology in mass data storage and computing.The large data analysis process constructed in this paper is implemented with open source architecture.Among them,crawler technology is adopted for big data collection and Hadoop is selected for big data platform.With HDFS,HBase and Hive as a storage system,computing framework uses the graphs parallel computing framework,and the comprehensive use of Sqoop,Oozie,Elasticsearch and other technology to implement the relational database to data import and export of large data platform,automation of data collection and storage processes,and establish a full-text index.The network data is mostly text type data,which needs to be structured to facilitate the processing of the latter.First of all,each agricultural website's vegetable and fruit and wholesale market information naming standards and specifications are different,must be re-normalized.Second,the data granularity on each agricultural web site is different,and the data needs to be standardized.The influence factors,price trend prediction,spatial regional characteristics and distribution method of vegetable and fruit prices were analyzed.The factors affecting the price were analyzed by using the influence factors of vegetable and fruit price,and the influential factors of weather,per capita income and preferences were extracted.Then,these influencing factors are used to construct the price prediction analysis process for vegetables and fruit which is suitable for large data scenarios.Finally,the price forecast of vegetables and fruits was predicted by using the process.Use of the space location information for fruits and vegetables price spatial variance analysis it is concluded that the distribution of the fruits and vegetables origin,and user preferences,regional information demand and output of fruits and vegetables distribution scheme analysis get the decision-making information distribution.The normal operation of the prototype of fruit price information monitoring and analysis using big data technology is proved to be feasible in this paper.After verification,the results of the prototype analysis are practical and the results are reliable.