论文部分内容阅读
证据权方法是目前最常用的信息综合方法之一,广泛应用于矿产资源定量预测与评价.然而,它要求变量间相互独立,地质上很难满足这一条件.如何削弱条件不独立对证据权预测结果的影响,已成为当前数学地球科学研究的热点.解决该问题的途径之一是对传统证据权模型进行校正,比如采取加权的方法对原证据权模型计算的证据权重进行修正,以便消除非条件独立性的影响.对近期提出的多种加权证据权模型进行了系统的对比研究,基于同样的应用实例和实验方案,对不同方法的应用效果进行了比较,结果表明,各种加权证据权模型均可不同程度地削弱证据图层条件不独立性的影响,其中,基于逻辑回归的加权证据权模型优于其他加权方法.
The method of evidence right is one of the most commonly used methods of information synthesis and is widely used in the quantitative prediction and evaluation of mineral resources.However, it requires that the variables are independent of each other and it is difficult to meet the requirement in geology. How to weaken the conditions is not independent of the right of evidence The impact of forecasting results has become a hot topic in current mathematical geosciences research.One of the ways to solve this problem is to correct the traditional model of evidence right, for example, to take the weighted method to correct the weight of evidence calculated by the original model of evidence in order to eliminate The impact of non-conditional independence on a variety of recent proposed weighting weight model of a systematic comparative study, based on the same application examples and experimental programs, the effectiveness of different methods were compared, the results show that a variety of weighted evidence The weight model can reduce the influence of the independency of the evidence layer condition to some extent. Among them, the weighting weight model based on logistic regression is superior to other weighting methods.