【摘要】 通过提取陶瓷样品对瓷器鉴定的影响因素,采用了基于欧氏距离分析法,得到了各个因素的权重,从而能够确定影响因素的关键因素.通过关键因素作为神经网络模型输入变量,建立基于提取权重的概率神经网络算法.实例分析表明:算法通过提取权重能够提高分类准确度.在大样本实例分析中,算法比其他统计分析和相似度算法具有更快的收敛速度,并能够应用到其它数据处理中,具有广泛的适用性.更多还原

【Abstract】 By extracting the influencing factors of identification about ceramic samples in the official,analysising based on Euclidean distance,we gained the weight of various factors, and determining the key factors of the porcelain identification.we established the probabilistic neural network algorithm of basing on extracting weight through using key factors as input variables.The example analysis shows that this algorithm has more accurate in classification results.the algorithm is faster in convergence speed th$n other statistic analysis and similarity algorithm in the large sample instance,and having a wide range of applicability to be applied to other data processing.更多还原