基于加速遗传算法的投影寻踪聚类评价模型研究与应用

Study and Application of PPC Model Based on RAGA in the Clustering Evaluation

针对农业生产力综合评价这类高维指标体系决策问题,采用降维技术:投影寻踪分类模型,利用基于实数编码的加速遗传算法优化其投影方向,将多维数据指标(样本评价指标)转换到低维子空间,根据投影函数值的大小评价出样本的优劣,从而做出决策。该模型最大限度地避免了传统评判中权重取值的人为干扰,评价结果更为准确客观,为农业生产力综合评价决策及其它评判决策问题提供一条新的方法与思路。

Through applying Projection Pursuit Cluster （PPC） model combined with Real coding based Accelerating Genetic Algorithm （RAGA） in the comprehensive decision-making of agricultural production capacity, the authors turn multidimension data into low dimension space. So the optimum projection direction can stand for the best influence to the collectivity. Thus, the value of projection function can evaluate each item good or not. The RAGA--PPC model can furthest avoid jamming of weight in the traditional method of multi-dimension data, and obtain better result. The model provides a new method and thought for readers who engaged in the comprehensive decision-maklng of agricultural production capacity and other relative study.