采用多蜂群协同演化算法的服装流行色预测

Hierarchical multi-hive bee colony algorithm for fashion color prediction

针对现有服装流行色定量预测方法存在的精度缺陷,借鉴了复杂适应系统理论和生物学协同进化思想,提出一种基于层次协同演化机制的多蜂群协同优化算法。将该算法应用于人工神经网络权值配置问题上,通过拟合曲线进行测试。结果表明该算法可提升神经网络模拟目标问题的精度。使用PANTONE发布的2007—2016年的国际春夏女装流行色定案,建立了春夏女装流行色色相的预测模型;对该模型中改进的神经网络进行训练,分析不同隐含层节点个数对色相预测精度的影响;预测2017年女装流行色色相,并将预测结果与PANTONE官方发布结果进行对比。结果表明该方法与其他方法相比提高了预测结果的精准度。

Aiming at the accuracy defect of current prediction method on fashion color,a hierarchical multi-hive artificial bee colony algorithm（ HABC） was proposed based on complex adaptive system theory and biological coordinated evolutionary thought. HABC was applied in calculating the weight values of network and tested by fitting curves. The results show that it can improve the accuracy of neural network simulation of the target problem. Using the data of women international spring and summer clothing fashion color in the years from 2007 to 2016 released by PANTONE,a prediction model of fashionable color hue is established. The influence of nodes number in different hidden layers on hue prediction accuracy is analyzed. By comparing the prediction results with that issued by PANTONE official,it is shown that the proposed method improves the prediction accuracy compared with the other methods.