论文部分内容阅读
为准确把握消费者对产品造型意象的认知规律,将支持向量机模型引入消费者的意象评价中。以数控机床为例,调查数控机床的意象得分,然后对数控机床的造型要素进行分解,最后将获得的数据代入粒子群算法优化支持向量机中进行学习,获得产品意象评价的数据模型。为比较BP神经网络、交叉验证法支持向量机、粒子群算法支持向量机这3种方法的准确性,抽选出5组没有在前面进行学习的数据进行测试。实验结果表明,粒子群算法支持向量机模型预测出的意象评价平均值与问卷调查所得平均值比其他两种方法更为接近,可以较好地用于预测消费者对产品造型的意象评价。
In order to accurately grasp the consumer’s cognitive rules of product styling and imagery, the SVM model is introduced into the consumer’s image evaluation. Taking CNC machine tool as an example, this paper investigates the image score of CNC machine tools, then decomposes the modeling elements of CNC machine tool, and finally substitutes the obtained data into PSO-SVM to obtain the data model of product image evaluation. In order to compare the accuracy of the three methods of BP neural network, cross-validation SVM and PSO SVM, five groups of data that have not been studied in the past are tested. The experimental results show that the average value of the image evaluation predicted by PSO model is closer to the average value of the questionnaire survey than the other two methods, which can be better used to predict consumers’ image evaluation of product modeling.