论文部分内容阅读
为了消除和减弱当证据层不满足条件独立性假设时对预测结果产生的影响,提出了逐步证据权模型和加权证据权模型.加权证据权模型通过对logit模型进行修改,对各个证据层给予一定的权重,以调整由于证据层与其他证据层的条件相关性对模型的影响;逐步证据权模型是将证据层按照一定的顺序逐步加入到模型中,在加入到模型的过程中依次用已经获得的后验概率作为模糊训练层的方法.以个旧锡铜多金属矿产资源预测为例,应用4种证据权模型的后验概率进行异常圈定,结果表明两种新的模型对减弱证据层不满足条件独立性假设所产生的影响是有效的.
In order to eliminate or weaken the influence of the evidence layer on the prediction results when the conditional independence assumption is not satisfied, a stepwise evidence weight model and a weighting weight model are proposed. The weighting weight model modifies the logit model to give certain levels of evidence In order to adjust the influence of the conditional correlation of the evidence layer and other evidence layers on the model. The progressive evidence weight model gradually adds the evidence layer to the model according to a certain sequence. In the process of adding to the model, Of the posterior probability as a fuzzy training layer method.According to the prediction of Gejiu tin copper polymetallic mineral resources, the paper uses the posterior probability of the four kinds of evidence weight models to delineate the abnormalities. The results show that the two new models are not satisfied with the weakened evidence layer The impact of the conditional independence assumption is valid.