Research On Key Information Mining Of Agricultural Products Big Data Based On E-commerce Platform

With the development of science and technology,E-commerce is developing rapidly as a digital-based platform.It is accompanied by the behavior of consumers and enterprises,and generates a large amount of complex data in real time,which often contain great business value.Agricultural products are the material information for basic living needs.Under the huge demand market,E-commerce will effectively unify the time and space,so that China's agricultural products economy has re-emerged vitality.This article mines the potentially valuable information of the big data of agricultural products on the E-commerce platform,and applies the mining analysis results to real life,in an effort to provide service support for E-commerce enterprises to realize effective management and precise marketing.On the basis of considering the wide distribution of agricultural products,the universality of consumers' preferences for agricultural products,and the availability of data,apples were selected as the research object of agricultural products in this paper.Through the exploration and analysis and preprocessing of Apple's big data,the clustering analysis and outlier detection and mining tasks of the obtained high-quality target data set are studied.K-Means clustering algorithm is adopted to conduct two-dimensional and multidimensional clustering of apple's big data,clustering of Apple's sales stores can be realized from different angles.Sales stores with similar characteristics are found,and their characteristics are described in detail to sum up.According to the apple big data obtained in this paper,the analysis of two-dimensional and multi-dimensional clustering results shows that the effect of two-dimensional clustering is better than that of multi-dimensional clustering.Outlier detection mainly excavates the apple sales stores where there are brush sales,purchase fans and deliberately improve the store score.During the detection,find the factors that are strongly related to the abnormal phenomenon,and use the method based on statistics and clustering to mine abnormal stores.Finally,the abnormal apple sales stores were found and displayed in the paper.Based on the E-commerce platform,the big data of agricultural products is mined and researched.The agricultural product sales shops are grouped from different dimensions and perspectives,and the agricultural product sales shops with abnormal behaviors are detected.Theoretically speaking,this research has established a more accurate and effective research method for mining big data of E-commerce agricultural products,which has certain reference significance;From the perspective of practical application,the relevant results obtained by this research can effectively screen and manage the agricultural product sales stores that are settled by the power supplier companies,ensure that consumers on the E-commerce platform enjoy a fair and transparent consumption environment,and realize the precise planting of farmers,and thus have important practical significance to promote the targeted poverty alleviation pace of rural E-commerce.