基于大数据的最优招标方案选取算法

Optimal bidding scheme selection algorithm based on big data

传统招标方案选取算法通过建立指标属性矩阵和直觉模糊线性评价模型进行最优招标方案选取,没有考虑大数据处理上的时间性,导致选取结果效率低,准确性差。为此提出基于大数据的最优招标方案选取算法,对全部招标方案进行聚类,通过大数据抽样加快对招标方案的大数据处理进程;选用Single簇质心方位;利用均值更新方法对大数据自然簇质心的方位进行修改以确定大数据自然簇质心的实质方位,在此基础上对招标方案集进行划分聚类。实现招标方案的聚类后,构建投影寻踪模型进行最优招标方案的选取。实验结果表明,所提算法聚类误差低于7%,准确性高达93%,并且在计算速度方面具有较大优势。

The traditional bidding scheme selection algorithm makes the optimal bidding scheme selection by establishing the index attribute matrix and intuitive fuzzy linear evaluation model,and does not consider the timing of big data processing,resulting in low efficiency and poor accuracy of the selected results.Therefore,an optimal bidding scheme selection algorithm based on big data is put forward.All the bidding schemes are clustered.The big data processing process of bidding schemes is accelerated by means of big data sampling.The Single method is selected to cluster the results of big data sampling,so as to determine the centroid orientation for the natural cluster of big data.The mean value updating method is adopted to modify the centroid orientation for the natural cluster of big data,so as to determine the actual centroid orientation for the natural cluster of big data.On this basis,the classification and clustering of the bidding schemes are conducted.The projection tracing model is constructed to select the optimal bidding scheme.The experimental results show that the proposed algorithm has a clustering error of less than 7%,accuracy of as high as 93%,and a great advantage in computing speed.