基于互联网+数据挖掘的农业数据平台设计

随着农业数据规模日益增加,相应的、有针对性的数据挖掘平台设计成为迫切需要.该文设计了基于互联网+数据挖掘的农业数据平台,包括交互层、功能层、数据层3个层次.功能层是平台的核心,负责数据预处理、数据挖掘、挖掘结果展示等任务.针对Apriori算法进行了改进,通过频度集合、支持度集合、地址集合的联合使用,避免了重复扫描和频度冗余,进一步提升了挖掘算法的效率.以水稻生长中的二化螟虫害问题为研究对象,展开平台性能的验证性实验.实验结果表明:4种因素对二化螟爆发影响的强弱顺序为5月份降水最大,其次是5月份的温度,再次是2月份的温度,最后是2月份的降水.

Design of Agricultural Data Platform Based on Internet + Data Mining

With the increasing scale of agricultural data, the design of corresponding data mining platform has become an urgent problem. An agricultural data platform has been designed in this paper based on Internet + data mining, which consists of three levels:interaction layer, function layer and data layer. Functional layer is the core of the platform, responsible for data preprocessing, data mining, mining results display and other tasks. The Apriori algorithm is improved. By using frequency set, support set and address set together, it avoids repeated scanning and frequency redundancy, and further improves the efficiency of mining algorithm. Taking the problem of rice stem borer pests in rice growth as the research object, the validation experiment of platform performance was carried out. The results show that the order of four factors influencing the outbreak of stem borer was the precipitation in May, followed by the temperature in May, February and February.